HEALTHINF 2018 Abstracts


Full Papers
Paper Nr: 12
Title:

Temporal Conformance Analysis and Explanation on Comorbid Patients

Authors:

Luca Piovesan, Paolo Terenziani and Daniele Theseider Dupré

Abstract: The treatment of comorbid patients is one of the main challenges of modern health care, and many Medical Informatics approaches have been devoted to it in the last years. In this paper, we propose the first approach in the literature that analyses the conformance of execution traces with multiple Computer-Interpretable Guidelines (CIGs), as needed in the treatment of comorbid patients. This is a fundamental task, to support physicians in an a-posteriori analysis of the treatments that have been provided. Notably, the conformance problem is very complex in this context, since CIGs may have negative interactions, so that in specific circumstances full conformance to individual CIGs may be dangerous for patients. We thus complement our conformance analysis with an explanation approach, aimed at justifying deviations in case they can be explained in terms of interaction management, e.g., some possible undesired interaction has been avoided. Our approach is based on Answer Set Programming, and, to face realistic problems, devotes specific attention to the temporal dimension.

Paper Nr: 15
Title:

The Privacy Management Platform - An Enabler for Device Interoperability and Information Security in mHealth Applications

Authors:

Christoph Stach, Frank Steimle and Bernhard Mitschang

Abstract: Chronic diseases are on the rise. Afflicted patients require persistent therapy and periodic screenings. This causes high treatment costs and overburdened physicians. Innovative approaches that enable patients to perform treatment methods on their own are badly needed. Telemedical approaches with the aid of modern Smartphones connected to medical devices (the so-called mHealth) can be the answer. However, mHealth apps face two key challenges, namely device interoperability and information security. In this paper, we describe how the Privacy Management Platform (PMP) and its extendable Resources can contribute to these challenges. Therefore, we analyze a real-world mHealth app and derive generic functional units, each realizing a certain task recurring frequently within mHealth apps, e.g., metering, data storage, or data transmission. For each functional unit we provide a PMP Resource, enabling both, device interoperability and information security. Finally, we revise the analyzed mHealth app using the Resources in order to evaluate our approach.

Paper Nr: 23
Title:

Stair Climb Power Measurements via Inertial Measurement Units - Towards an Unsupervised Assessment of Strength in Domestic Environments

Authors:

Sandra Hellmers, Tobias Kromke, Lena Dasenbrock, Andrea Heinks, Jürgen M. Bauer, Andreas Hein and Sebastian Fudickar

Abstract: In order to initiate interventions at an early stage of functional decline and thus, to extend independent living, the early detection of changes in functional ability is important. The Stair Climb Power Test (SCPT) is a standard test in geriatric assessments for strength as one of the essential components of functional ability. This test is also well suited for regular and frequent power measurements in daily life since the activity of climbing stairs is usually frequently performed. We introduce an automated assessment of the SCPT based on inertial measurement units (IMU) in a study of 83 participants aged 70-87 years. For power evaluations of the lower extremities, the activity of climbing stairs was automatically classified via machine learning and the power was calculated based on the test duration and covered height. Climbing stairs was correctly classified in 93% of the cases. We also achieved a good correlation of the power calculations with the conventional stop watch measurements with a mean deviation of 2.35%. The system’s sensitivity to detect the transition towards frailty has been confirmed. Furthermore, we discussed the general suitability of the automated stair climb power algorithm in unsupervised, standardized home-assessments.

Paper Nr: 26
Title:

SClinico: Usability Study

Authors:

João Pavão, Rute Bastardo, Marta Covêlo, Luís Torres Pereira, Paula Oliveira, Catarina Pedrosa, Ana Silva, Victor Costa, Ana Isabel Martins, Alexandra Queirós and Nelson Pacheco Rocha

Abstract: The use of electronic health records (EHR) to support clinical practices is widespread worldwide, due to the need to optimize health care delivery. Therefore, the usability assessment of EHR systems is crucial. The objective of this study was to perform a qualitative and quantitative assessment of the usability of SClinico, the most used EHR system within the Portuguese National Health Service. This observational study to assess SClinico usability took place in several clinical services of the Centro Hospitalar de Trás-os-Montes e Alto Douro. The results show that SClinico has some usability issues that influence the clinical practice and, therefore, need to be improved.

Paper Nr: 27
Title:

Interoperability in Pervasive Health: Is It Tackled as a Priority?

Authors:

Ana Dias, Ana Isabel Martins, Alexandra Queirós and Nelson Pacheco Rocha

Abstract: For the electronic health record (EHR) to be considered a true clinical decision support system, it must be possible to access and integrate the patients’ clinical information collected throughout their lives, guaranteeing up-to-date, safe and congruent information, immediately accessible at the place of care. Moreover, there is a considerable capacity to develop and manufacture personal health devices (PHD) highly integrated and miniaturized, which facilitate the home monitoring of patients with chronic diseases. Since the information collected by PHD should be integrated in existing EHR, interoperability is an essential requirement of eHealth to allow the integration of care into a diversity of settings and care providers. The purpose of this systematic review was to identify and analyse references related to the topic of home monitoring that reveal an explicit concern with interoperability requirements. Regarding the results and considering the initial 2778 references, only 2% (61 references) explicitly mentioned interoperability issues and, within these 61 references, only eight reported end-to-end solutions that can be integrated and usable in care service provision. Therefore, the issue of interoperability of PHD, both semantic and technological, a priority for the establishment of a remote patient monitoring solution market, is discussed in this review.

Paper Nr: 39
Title:

A General Framework for the Distributed Management of Exceptions and Comorbidities

Authors:

Alessio Bottrighi, Luca Piovesan and Paolo Terenziani

Abstract: In the last decades, many different computer-assisted management systems for Computer Interpretable Guidelines (CIGs) have been developed. While CIGs propose a “standard” evidence-based treatments of “typical” patients, exceptions may arise, as well the need to cope with comorbidities. The treatment of deviation from “standard” execution has attracted a lot of attention in the recent literature, but the approaches proposed are focused on the treatment either of exceptions or of comorbities. However, this is a clear limitation, since during a CIG execution, both these issues can occur. In this paper, we propose the first approach which supports the integrated treatment of both exceptions and comorbidities. To achieve such a goal, we propose a modular client-server architecture supporting the concurrent execution of multiple guidelines. The architecture proposed has been designed as a further layer building upon “traditional” execution engines for a single CIG. Thus, our methodology is general and can be used to extend the CIG systems in the literature. Finally, we describe our approach in action on a case study, in which a comorbid patient is treated for Peptic Ulcer and for deep Venous Thrombosis and, during the treatment, she manifests a heart failure.

Paper Nr: 42
Title:

Cross-cohort Evaluation of Machine Learning Approaches to Fall Detection from Accelerometer Data

Authors:

Aneta Lisowska, Alison O'Neil and Ian Poole

Abstract: Falls in seniors can lead to serious physical and psychological consequences. A fall detector can allow a fallen person to receive medical intervention promptly after the incident. The accelerometer data from smartphones or wearable devices can be used to detect falls without serious privacy intrusion. Common machine learning approaches to fall detection include supervised and novelty based methods. Previous studies have found that supervised methods have superior performance when tested on participants from the population cohort resembling the one they were trained on. In this study, we investigate if the performance remains superior when they are tested on a distinctly different population cohort. We train the supervised algorithms on data gathered using a wearable Silmee device (Cohort 1) and test on smartphone data from a publicly available data set (Cohort 2). We show that the performance of the supervised methods decreases when they are tested on distinctly different data, but that the decrease is not substantial. Novelty based fall detectors have better performance, suggesting that novelty based detectors might be better suited for real life applications.

Paper Nr: 46
Title:

Towards an Agent-based Model to Monitor Epidemics and Chronic Diseases in DR Congo

Authors:

Jean-Claude Tshilenge Mfumu, Annabelle Mercier, Christine Verdier and Michel Occello

Abstract: Many contagious diseases occurred around sub-Saharan countries in the last decade due to the inefficiency of health structures to anticipate disease outbreaks. In a huge poorly-infrastructured country such as The Democratic Republic of Congo (DRC) with insufficient health staff and laboratory facilities, to provide quick response to an urgent case of epidemic is challenging especially facing the development of its rural areas. As DRC’s Health System has three levels (peripheral, regional and national levels), from producing health data at peripheral to national level that takes the decision, it can take time resulting in the spread of disease. The lack of communication between health centers and laboratory facilities in the same health zone does not contribute to regional riposte. This paper proposes to face this problem using an agent-centered approach to study through simulation how to improve the process. An experiment is described by agentifying two health zones on the same regional level to show how it can reduce the decision time.. It consists of 2 peripheral coordination offices, 2 labs and 2 health zones the former with 12 health centers and the latter with 20 health zones. The interaction between these agents will provide a first model to be compared with the current system in other to reduce decision time.

Paper Nr: 50
Title:

A Remote Home Monitoring System to Support Informal Caregivers of People with Dementia

Authors:

Stefan Lentelink, Monique Tabak, Boris van Schooten, Dennis Hofs, Harm op den Akker and Hermie Hermens

Abstract: Informal caregivers of people with dementia have a high risk of becoming overburdened. Health informatics for aging in place can provide them support by deploying unobtrusive remote home monitoring systems to assess real-time events and monitor changes in the behavior of the person with dementia (PwD). In this paper, we describe the concept, development, and evaluation of an intelligent remote Home Monitoring System (HMS) that provides support to informal caregivers by giving key information related to the health and independent living of the PwD. The HMS consists of a Sensor System that monitors low-level behaviors of the PwD, a Decision Support System that translates this into high-level behaviors, and a connected Smartphone Application that allows the caregiver to receive notifications, review behavioral information at a glance, and facilitates the collaborative care process between informal caregivers. The final HMS prototype was evaluated and scored high in terms of usability and quality of the Smartphone Application. The Sensor System showed no significant flaws during testing, and the Decision Support System is considered a viable proof of concept. The next step is to evaluate the HMS in a real-life setting in terms of offering peace of mind and reducing the burden of care.

Paper Nr: 51
Title:

Predicting Future Antibiotic Susceptibility using Regression-based Methods on Longitudinal Massachusetts Antibiogram Data

Authors:

M. L. Tlachac, Elke Rundensteiner, Kerri Barton, Scott Troppy, Kirthana Beaulac and Shira Doron

Abstract: Antibiotic resistance evolves alarmingly quickly, requiring constant reevaluation of resistance patterns to guide empiric treatment of bacterial infections. Aggregate antimicrobial susceptibility reports, called antibiograms, are critical for evaluating the likelihood of effectiveness of antibiotics prior to the availability of patient specific laboratory data. Our objective is to analyze the ability of the methods to predict antimicrobial susceptibility. This research utilizes Massachusetts statewide antibiogram data, a rich dataset composed of average percent susceptibilities of 10 species of bacteria to a variety of antibiotics collected by the Massachusetts Department of Public Health from over 50 acute-care hospitals from 2002 to 2015. First, we improved data quality by implementing data filtering strategies. We then predicted up to three future years of antibiotic susceptibilities using regression-based strategies on nine previous years of data. We discovered the same prediction methodology should not be utilized uniformly for all 239 antibiotic-bacteria pairs. Thus, we propose model selection strategies that automatically select a suitable model for each antibiotic-bacteria pair based on minimizing those models' mean squared error and previous year's prediction error. By comparing the predictions against the actual mean susceptibility, our experimental analysis revealed that the model selectors based on the predictions of the previous performed best.

Paper Nr: 54
Title:

Introduction of a Bayesian Network Builder Algorithm - Personalized Infectious Disease Risk Prediction

Authors:

Retno Aulia Vinarti and Lucy Hederman

Abstract: We introduce an algorithm for auto-generating a Bayesian Network (BN) structure from a knowledge-base represented as an ontology with rules. The ontology and rules represent the assumptions of infectious disease risk in the epidemiology domain. The resulting BN will be the computational model for an infectious disease risk prediction service. The BN structure consists of one child node, to represent the chosen infectious disease, with multiple parent nodes to represent the contexts which affect infection risk. Thus, this BN generation algorithm is constrained to a relatively simple structure. The algorithm generates a BN using the API of BN modeler software, Netica-J. We evaluate two aspects of the generated BN: the network structure and the conditional probability tables (CPTs). The validation result shows that the algorithm generates an isomorphic BN compared with the ontology and the CPTs are populated with consistent ratios from epidemiological rules. Furthermore, the generated BN has resulted in a personalized infectious disease risk prediction based on the personal attributes and their environments.

Paper Nr: 58
Title:

A Methodology to Perform Semi-automatic Distributed EHR Database Queries

Authors:

Olga Fajarda, Luis Bastião Silva, Peter R. Rijnbeek, Michel Van Speybroeck and José Luis Oliveira

Abstract: The proliferation of electronic health databases has resulted in the existence of a wide collection of diversified clinical digital data. These data are fragmented over dispersed databases in different clinical silos around the world. The exploration of these electronic health records (EHRs) is essential for clinical and pharmaceutical research and, therefore, solutions for secure sharing of information across different databases are needed. Although several partial solutions have been proposed over the years, data sharing and integration has been hindered by many ethical, legal and social issues. In this paper, we present a methodology to perform semi-automatic queries over longitudinal clinical data repositories, where every data custodian maintains full control of data.

Paper Nr: 62
Title:

Automated Measurement of Adherence to Traumatic Brain Injury (TBI) Guidelines using Neurological ICU Data

Authors:

Anthony Stell, Ian Piper and Laura Moss

Abstract: Using a combination of physiological and treatment information from neurological ICU data-sets, adherence to traumatic brain injury (TBI) guidelines on hypotension, intracranial pressure (ICP) and cerebral perfusion pressure (CPP) is calculated automatically. The ICU output is evaluated to capture pressure events and actions taken by clinical staff for patient management, and are then re-expressed as simplified process models. The official TBI guidelines from the Brain Trauma Foundation are similarly evaluated, so the two structures can be compared and a quantifiable distance between the two calculated (the measure of adherence). The methods used include: the compilation of physiological and treatment information into event logs and subsequently process models; the expression of the BTF guidelines in process models within the real-time context of the ICU; a calculation of distance between the two processes using two algorithms (“Direct” and “Weighted”) building on work conducted in the business process domain. Results are presented across two categories each with clinical utility (minute-by-minute and single patient stays) using a real ICU data-set. Results of two sample patients using a weighted algorithm show a non-adherence level of 6.25% for 42 mins and 56.25% for 708 mins and non-adherence of 18.75% for 17 minutes and 56.25% for 483 minutes. Expressed as two combinatorial metrics (duration/non-adherence (A) and duration * non-adherence (B)), which together indicate the clinical importance of the non-adherence, one has a mean of A=4.63 and B=10014.16 and the other a mean of A=0.43 and B=500.0.

Paper Nr: 65
Title:

A Rule-based Method Applied to the Imbalanced Classification of Radiation Toxicity

Authors:

Juan L. Domínguez-Olmedo, Jacinto Mata, Victoria Pachón and Jose L. Lopez-Guerra

Abstract: This paper describes a rule-based classifier (DEQAR-C), which is set up by the combination of selected rules after a two-phase process. In the first phase, the rules are generated and sorted for each class, and then a selection is performed to obtain a final list of rules. A real imbalanced dataset regarding the toxicity during and after radiation therapy for prostate cancer has been employed in a comparison with other predictive methods (rule-based, artificial neural networks, trees, Bayesian and logistic regression). DEQAR-C produced excellent results in an evaluation regarding several performance measures (accuracy, Matthews correlation coefficient, sensitivity, specificity, precision, recall and F-measure) and by using cross-validation. Therefore, it was employed to obtain a predictive model using the full data. The resultant model is easily interpretable, combining three rules with two variables, and suggesting conditions that are mostly confirmed by the medical literature.

Paper Nr: 77
Title:

Early Prediction of MRSA Infections using Electronic Health Records

Authors:

Thomas Hartvigsen, Cansu Sen, Sarah Brownell, Erin Teeple, Xiangnan Kong and Elke Rundensteiner

Abstract: Despite eradication efforts, Methicillin-resistant Staphylococcus aureus (MRSA) remains a common cause of serious hospital-acquired infections (HAI) in the United States. Electronic Health Record (EHR) systems capture MRSA infection events along with detailed patient information preceding diagnosis. In this work, we design and apply machine learning methods to support early recognition of MRSA infection by estimating risk at several time points during hospitalization. We use EHR data including on-admission and throughout-stay patient information. On-admission features capture clinical and non-clinical information while throughout-stay features include vital signs, medications, laboratory studies, and other clinical assessments. We evaluate prediction accuracy achieved by core Machine Learning methods, namely Logistic Regression, Support Vector Machine, and Random Forest classifiers, when mining these different types of EHR features to detect patterns predictive of MRSA infection. We evaluate classification performance using MIMIC III – a critical care data set comprised of 12 years of patient records from the Beth Israel Deaconess Medical Center Intensive Care Unit in Boston, MA. Our methods can achieve near-perfect MRSA prediction accuracies one day before documented clinical diagnosis. Also, they perform well for early MRSA prediction many days in advance of diagnosis. These findings underscore the potential clinical applicability of machine learning techniques.

Paper Nr: 78
Title:

Mobile-based Risk Assessment of Diabetic Retinopathy using a Smartphone and Adapted Ophtalmoscope

Authors:

Simão Felgueiras, João Costa, João Gonçalves and Filipe Soares

Abstract: The large prevalence of diabetes in the global population is associated with an increasing number of Diabetic Retinopathy cases. This disease is associated with a progressive risk of blindness, due to physiological changes that affect the retina. Since most of the progression is asymptomatic and late stage damage is often irreversible, there is a large incentive to implement effective methodologies that allow large scale screening of the diabetic population. In this work, a research study of a mobile approach for the assessment of Diabetic Retinopathy was conducted, by analyzing 80 patients already being followed for ophthalmological care. A smartphone-based fundus imaging system was used to acquire images of the retina during the normal clinical workflow in a Central Hospital in Portugal. Relevant images were automatically analyzed by a Decision Support System (DSS) based on computer vision methods. The results were obtained for ground-truth correlation as well as time impact of this novel system. Our conclusions support that the DSS is highly sensitive in detecting pathological information on images, after a dedicated quality image filtering, and the acquisition procedure has minimal adverse impact in the clinical setting.

Paper Nr: 80
Title:

One Size Does Not Fit All: An Ensemble Approach Towards Information Extraction from Adverse Drug Event Narratives

Authors:

Susmitha Wunnava, Xiao Qin, Tabassum Kakar, Xiangnan Kong, Elke A. Rundensteiner, Sanjay K. Sahoo and Suranjan De

Abstract: Recognizing named entities in Adverse Drug Reactions narratives is a fundamental step towards extracting valuable patient information from unstructured text into a structured thus actionable format. This then unlocks advanced data analytics towards intelligent pharmacovigilance. Yet existing biomedical named entity recognition (NER) tools are limited in their ability to identify certain entity types from these domain-specific narratives and result in significant performance differences in terms of accuracy. To address these challenges, we propose an ensemble approach that integrates a rich variety of named entity recognizers to procure the final result. First, one critical problem faced by NER in the biomedical context is that the data is highly skewed. That is, only 1% of words belong to a certain medical entity type, such as, the reason for medication usage compared to all other non-reason words. We propose a balanced, under-sampled bagging strategy that is dependent on the imbalance level to overcome the class imbalance problem. Second, we present an ensemble of heterogeneous recognizers approach that leverages a novel ensemble combiner. Our experimental results show that for biomedical text datasets: (i) a balanced learning environment along with an Ensemble of Heterogeneous Classifiers constantly improves the performance over individual base learners and, (ii) stacking-based ensemble combiner methods outperform simple Majority Voting by 0.30 F-measure.

Paper Nr: 81
Title:

Identifying Characteristic Physiological Patterns of Parkinson's Disease Sufferers using Sample Entropy of Pulse Waves

Authors:

Mayumi Oyama-Higa, Tokihiko Niwa, Wenbiao Wang and Yoshifumi Kawanabe

Abstract: In this study, we identify characteristic physiological patterns of Parkinson’s disease patients, through analysis of the data of their pulse waves. We find that the sample entropy values of pulse waves, with certain parameters fix (In this case, we define the sample entropy value as “border of Parkinson entropy”, or BPE), is statistically different between Parkinson’s disease sufferers and healthy individuals. In addition, values of the largest Lyapunov exponent computed from the same data are also analysed, and significant difference between the two groups are observed. At the end, we describe an Android tablet that we developed for real-time measurement and analysis of BPE.

Paper Nr: 88
Title:

Usability of a New eHealth Monitoring Technology That Reflects Health Care Needs for Older Adults with Cognitive Impairments and Their Informal and Formal Caregivers

Authors:

Fatma Cossu-Ergecer, Marit Dekker, Bert-Jan F. van Beijnum and Monique Tabak

Abstract: The aim of this study was to evaluate an eHealth monitoring application (HELMA) that provides insight in the health status of older adults with cognitive impairments (CI) independently living at home and their caregivers. A mixed-method approach was used to collect data on Usability (System Usability Scale) and Actual Use (Log data). Besides, a subgroup of participants were randomly selected and interviewed about their experiences with HELMA (Ease of Use, Perceived Usefulness, Behavioural Intention to Use and Attitude). Fifty-four older adults, fifteen formal and fourteen informal caregivers participated in this study. Results showed that HELMA is a useful supplement in the current care for older adults with cognitive impairments. The average SUS score of HELMA of formal caregivers indicated “good” usability. The questions of HELMA are clear. However, older adults lacked digital skills to use HELMA by themselves. Most of the participants (80%) used HELMA according to protocol, for a minimum of 4 weeks. The attitude towards willingness to learn and to use a technology were negative for almost all older adults. More attention to different implementation strategies is needed to increase the eHealth literacy of older adults with CI, to improve independent use of HELMA in the future.

Paper Nr: 102
Title:

Supporting Multiple Agents in the Execution of Clinical Guidelines

Authors:

Alessio Bottrighi, Luca Piovesan and Paolo Terenziani

Abstract: Clinical guidelines (GLs) exploit evidence-based medicine to enhance the quality of patient care, and to optimize it. To achieve such goals, in many GLs different agents have to interact and cooperate in an effective way. In many cases (e.g. in chronic disorders) the GLs recommend that the treatment is not performed/completed in the hospital, but is continued in different contexts (e.g. at home, or in the general practitioner’s ambulatory), under the responsibility of different agents. Delegation of responsibility between agents is also important, as well as the possibility, for a responsible, to select the executor of an action (e.g., a physician main retain the responsibility of an action, but delegate to a nurse its execution). To manage such phenomena, proper support to agent interaction and communication must be provided, providing them with facilities for (1) treatment continuity (2) contextualization, (3) responsibility assignment and delegation (4) check of agent “appropriateness”. In this paper we extend GLARE, a computerized GL management system, to support such needs. We illustrate our approach by means of a practical case study.

Paper Nr: 110
Title:

The ATEN Framework for Creating the Realistic Synthetic Electronic Health Record

Authors:

Scott McLachlan, Kudakwashe Dube, Thomas Gallagher, Bridget Daley and Jason Walonoski

Abstract: Realistic synthetic data are increasingly being recognized as solutions to lack of data or privacy concerns in healthcare and other domains, yet little effort has been expended in establishing a generic framework for characterizing, achieving and validating realism in Synthetic Data Generation (SDG). The objectives of this paper are to: (1) present a characterization of the concept of realism as it applies to synthetic data; and (2) present and demonstrate application of the generic ATEN Framework for achieving and validating realism for SDG. The characterization of realism is developed through insights obtained from analysis of the literature on SDG. The development of the generic methods for achieving and validating realism for synthetic data was achieved by using knowledge discovery in databases (KDD), data mining enhanced with concept analysis and identification of characteristic, and classification rules. Application of this framework is demonstrated by using the synthetic Electronic Healthcare Record (EHR) for the domain of midwifery. The knowledge discovery process improves and expedites the generation process; having a more complex and complete understanding of the knowledge required to create the synthetic data significantly reduce the number of generation iterations. The validation process shows similar efficiencies through using the knowledge discovered as the elements for assessing the generated synthetic data. Successful validation supports claims of success and resolves whether the synthetic data is a sufficient replacement for real data. The ATEN Framework supports the researcher in identifying the knowledge elements that need to be synthesized, as well as supporting claims of sufficient realism through the use of that knowledge in a structured approach to validation. When used for SDG, the ATEN Framework enables a complete analysis of source data for knowledge necessary for correct generation. The ATEN Framework ensures the researcher that the synthetic data being created is realistic enough for the replacement of real data for a given use-case.

Paper Nr: 113
Title:

Local Multiple Sequence Alignment with Biclustering

Authors:

Wassim Ayadi

Abstract: Multiple Sequence Alignment (MSA) generally refers to cluster conserved subsequences. However, the choice of the clustering method can easily impact the quality of the alignment. In this context, we use the biclustering technique to generate a local multiple alignment which is independent of the types of biological sequences (DNA, RNA and proteins). Until now, the use of biclustering to solve the MSA problem is not well explored. In this paper, we present the Biclustering-based local MSA algorithm, called BiLMSA, that uses the bicluster enumeration approach to solve the problem of multiple sequence alignment. BiLMSA looks for aligning the maximum of blocks having the maximum relations with a set of sequences. BiLMSA was tested on proteins, RNA and DNA families. Our algorithm provides the best alignments compared to some of the best known algorithms and comparable to some others.

Short Papers
Paper Nr: 8
Title:

Exploring Quantified Self Attitudes

Authors:

Christel De Maeyer and Panos Markopoulos

Abstract: In recent years there is a growing optimism that health interventions may become more effective through the use of self-tracking. Related efforts are hampered by the short-lived compliance to self-tracking schemes. This paper examines the attitudes and motivations of self-trackers that could guide the design of self-tracking applications. Based on a questionnaire survey and follow up interviews a set of three personas of self trackers is proposed, in addition, design requirements are proposed for improving adherence to self-tracking technologies.

Paper Nr: 10
Title:

Renal Health - A New Tool for Chronic Kidney Disease - Application Development and a Proposal for Interventional Study

Authors:

Juliana Gomes Ramalho de Oliveira, José Eurico Vasconcelos Filho and Geraldo Bezerra da Silva Junior

Abstract: The aim of this study was to create an application for smartphones for chronic kidney disease (CKD). The development of the application was conducted in three phases: data collection, conception and development of an application called “Renal Health”. In the first phase, a literature review was conducted to ground the necessity of a tool to teach the general population about CKD and to give support to CKD patients in their treatment. Semi-structured interviews were then conducted with CKD patients (in hemodialysis or kidney transplant) and the general population to enhance out understanding of the main knowledge gaps about kidney disease. Individuals without CKD reported not knowing the disease (66.7%). Patients on hemodialysis reported difficulties with medication intake and diet (>50%). Transplanted patients had no problems with medication intake and showed to want more nutritional advices. After the development of the application, an usability test was done with CKD patients and specialists to evaluate its clarity and performance, and its acceptance was 89.6%. The use of Renal Health application can be an important tool for the general population, for knowledge acquisition, patients, health care workers, as well as patients’ family and caregivers of elderly and children patients.

Paper Nr: 16
Title:

ECG-derived Blood Pressure Classification using Complexity Analysis-based Machine Learning

Authors:

Monika Simjanoska, Martin Gjoreski, Ana Madevska Bogdanova, Bojana Koteska, Matjaž Gams and Jurij Tasič

Abstract: The recent advancement on wearable physiological sensors supports the development of real-time diagnosis in preventive medicine that demands various signal processing techniques to enable the extraction of the vital signs (e.g., blood pressure). Blood pressure estimation from physiological sensors data is challenging task that usually is solved by a combination of multiple signals. In this paper we present a novel complexity analysis-based machine-learning perspective on the problem of blood pressure class estimation only from ECG signals. We show that high classification accuracy of 96.68% can be achieved by extracting information via complexity analysis on the ECG signal followed by applying a stack of machine-learning classifiers. In addition, the proposed stacking approach is compared to a traditional machine-learning approaches and feature analysis is performed to determine the influence of the different features on the classification accuracy. The experimental data was gathered by daily monitoring of 20 subjects with two different ECG sensors.

Paper Nr: 19
Title:

Devices Used for Non-Invasive Tele Homecare for Cardiovascular Patients - A Systematic Literature Review

Authors:

Jessica van der Zweth, Marjan Askari, Marco Spruit and Christof van Nimwegen

Abstract: The aim of this Systematic Review (SLR) was to provide an overview of devices used for non-invasive tele health care by patients diagnosed with heart failure (HF). All English studies published in the past 10 years that focused on tele home care for coronary heart diseases, cardiac arrhythmia or heart failure patients were systematically searched in Scopus and Pubmed. Articles were selected if added value of tele-monitoring for these patients was studied. Selected titles and abstracts were screened to determine eligibility for further review. Types and number of devices per disease were then withdrawn and categorized for the three diseases. Eight devices were found in the literature to be used in non-invasive tele homecare for patients diagnosed with heart failure, of which weight scales and blood pressure monitors were most commonly used and are the most frequently occurring combination. The knowledge on which tele homecare devices are most commonly used gives insights into which devices are the best choice needed for patients with heart failure, where product suppliers and healthcare providers can respond to. The results create future directions for studying different aspects of tele homecare devices, such as usability aspects, which is an important factor for acceptance of telemedicine.

Paper Nr: 24
Title:

Relations of Morbidity Related Groups (MRG), ICD-10 Codes and Age and Gender Structure in Outpatient Treatment

Authors:

Reinhard Schuster, Thomas Ostermann, Marc Heidbreder and Timo Emcke

Abstract: A patient’s (basic) Morbidity Related Group (MRG) is defined by the drug class (first four characters of the international Anatomic Therapeutic Chemical [ATC] Classification System) with the highest costs per quarter with respect to a physician. The morbidity of a patient is thereby represented by the drug most important economically. We consider the relation of those case groups with diagnoses (ICD-10-GM) on the individual and group level. In analogy to the DRG Systems (Diagnosis-related group) a degree of severity with respect to age, multimorbidity and treatment intensity is defined. We compare multimorbidity and age structures of MRGs and ICD-10 using a distance measure given by the fraction of patients with respect to their MRG and ICD-10. Main diagnoses like in hospital treatment are not given in outpatient care. MRG classification data can be used in order to algorithmically construct an outpatient care equivalent. Individual MRG components as points in a vector space can be used to determine the ”biological age“ of groups of individuals with respect to in- or decreased morbidity.

Paper Nr: 36
Title:

On Detecting Chronic Obstructive Pulmonary Disease (COPD) Cough using Audio Signals Recorded from Smart-Phones

Authors:

Anthony Windmon, Mona Minakshi, Sriram Chellappan, Ponrathi R. Athilingam, Marcia Johansson and Bradlee A. Jenkins

Abstract: Chronic Obstructive Pulmonary Disease (COPD) is a lung disease that makes breathing a strenuous task with chronic cough. Millions of adults, worldwide, suffer from COPD, and in many cases, they are not diagnosed at all. In this paper, we present the feasibility of leveraging cough samples recorded using a smart-phone’s microphone, and processing the associated audio signals via machine learning algorithms, to detect cough patterns indicative of COPD. Using 39 adult cough samples evenly spread across both genders, that included 23 subjects infected with COPD and 16 Controls, not infected with COPD, our system, using Random Forest classification techniques, yielded a detection accuracy of 85:4% with very good Precision, Recall and FMeasures. To the best of our knowledge, this is the first work that designs a smart-phone based learning technique for detecting COPD via processing cough.

Paper Nr: 38
Title:

Sharing Genetic Data under US Privacy Laws

Authors:

Michael Reep, Bo Yu, Duminda Wijesekera and Paulo Costa

Abstract: Clinical medical practice and biomedical research utilize genetic information for specific purposes. Irrespective of the purpose of obtaining genetic material, methodologies for protecting the privacy of patients/donors in both clinical and research settings have not kept pace with rapid advances in genetics research. When the usage of genetic information is not predicated on the latest laws and policies, the result places all-important patient/donor privacy at risk. Some methodologies err on the side of overly stringent policies that may inhibit research and open-ended diagnostic activity, whereas an opposite approach advocates a high-degree of openness that can jeopardize patient privacy, inappropriately identify disease susceptibility of patients and their genetic relatives, and thereby erode the doctor-patient privilege. As a solution, we present a framework based on the premise that acceptable clinical treatment regimens are captured in workflows used by caregivers and researchers and therefore their associated purpose are inherent to and therefore can be extracted from these workflows. We combine these purposes with applicable consents that are derived from applicable laws and practice standards to ascertain the releasability of genetic information. Given that federal, state and institutional laws, rules and regulations govern the use, retention and sharing of genetic information, we create a three-level rule hierarchy to apply the laws to a request and auto-generate consents prior to releasing. Our hierarchy also identifies all pre-conditions that must be met prior to the genetic information release, any restrictions and constraints to be enforced after release, and the penalties that may be assessed for violating these terms. We prototype our system using open source tools, while ensuring that the results can be added to existing Electronic Medical Records (EMR) systems.

Paper Nr: 44
Title:

A Machine Translation Approach for Medical Terms

Authors:

Alejandro Renato, José Castaño, Pilar Ávila, Hernán Berinsky, Laura Gambarte, Hee Park, David Pérez, Carlos Otero and Daniel Luna

Abstract: We describe the task of translating clinical term descriptions from Spanish to Brazilian Portuguese. We build a statistical machine translation system (SMT) using in-domain parallel corpora and available machine learning tools. The performance of this SMT was compared with general purpose machine translation systems available online. We used different techniques to validate the result of the different systems, using reference domain terminology and the occurrence of translated descriptions in a corpus of medical scientific literature and in domain specific web pages. We also use two sets of 1000 description terms that were revised and checked by a Portuguese speaker. The performance of the SMT we built had very good preliminary results.

Paper Nr: 52
Title:

Extracting Information and Identifying Data Structures in Pharmacological Big Data using Gawk

Authors:

Reinhard Schuster, Timo Emcke, Martin Schuster and Thomas Ostermann

Abstract: In the past decades, health related data saw an increase in capture, storage and analysis of very large data sets, referred to as big data. In health services research, big data analysis is used by health policy makers to identify and forecast potential risk factors, causalities or hazards. However, big data due to its volume, its high variety of data types and its high velocity of data flow is often difficult to process. Moreover, big data also shows a high complexity of data structures. In such cases, gawk programming language is a powerful tool to work with by using structural elements such as associative arrays. This article aims at describing the use and interaction of gawk to extract information and identify data structures in pharmacological big data sets. In particular we aimed at showing its strength in combining it with Mathematica based on two examples of the prescription data for potentially inadequate medications for elderly patients and the creation of networks of physicians and drug related neighbourhood relations.

Paper Nr: 56
Title:

Length of Hospital Stay Prediction through Unorganised Turing Machines

Authors:

Luigi Lella and Ignazio Licata

Abstract: Length of hospital stay (LoS) prediction is one of the most important goals in Health Informatics, due to the fact that through this it is possible to optimize the management of health structure resources. In Italian local healthcare systems we are experimenting an health cost containment process and the minimization of care costs is considered an important objective to be achieved. For this reason we have tested several datamining models trained with hospital discharge data, capable to make accurate LoS predictions. In another work we have reached encouraging results by the use of unsupervised models which detect autonomously the subset of non-class attributes to be considered in these classification tasks. Here we are interested in studying also another intelligent data analysis model, the Turing unorganised A-type machine, that is capable to represent the acquired knowledge in a logic formalism. In other terms this solution can explain its predictions by the use of a set of self-acquired knowledge base rules.

Paper Nr: 61
Title:

A Modular Workflow Management Framework

Authors:

João Rafael Almeida, Ricardo Ribeiro and José Luis Oliveira

Abstract: Task management systems are crucial tools in modern organizations, by simplifying the coordination of teams and their work. Those tools were developed mainly for task scheduling, assignment, follow-up and accountability. Then again, scientific workflow systems also appeared to help putting together a set of computational processes through the pipeline of inputs and outputs from each, creating in the end a more complex processing workflow. However, there is sometimes a lack of solutions that combine both manually operated tasks with automatic processes, in the same workflow system. In this paper, we present a web-based platform that incorporates some of the best functionalities of both systems, addressing the collaborative needs of a task manager with well-structured computational pipelines. The system is currently being used by a European consortium for the coordination of clinical studies.

Paper Nr: 63
Title:

How to Design Game-based Healthcare Applications for Children? - A Study on Children’s Game Preferences

Authors:

A. F. A. de Vette, M. Tabak and M. M. R. Vollenbroek-Hutten

Abstract: Game-based design can be used to develop engaging health applications for children. This engagement can only be realised when design is tailored to their preferences. In this study we investigate game preferences of children and translate these into design recommendations. Game preferences of children aged 6 to 12 were assessed through a questionnaire. Outcomes were classified by means of the 7D framework which divides game content into seven linear domains. Significant differences in mean scores among demographic subgroups were explored. Sixty-five children participated (M=9 years, SD=0.24, 36 boys, 29 girls, 8 children with asthma). Data showed high preference for content in domains novelty (Mnovelty=63) and dedication (Mdedication=70). Analysis resulted in subdivision of scores based on gender, age and playing frequency. Striking differences in scores were found between boys and girls in discord (Mboys=62, Mgirls=19), intensity (Mboys=60, Mgirls=27), rivalry (Mboys=53, Mgirls=31) and threat (Mboys=64, Mgirls=25). To design games for children we recommend to stimulate curiosity by offering variation and discovery, to enable achievement, learning and social contact. A divergence in preferences for boys and girls must be regarded. Opposed to boys, girls may lose interest in games that have violent or scary content, that are mainly competitive or demand continuous effort.

Paper Nr: 64
Title:

Automatic Computation of Biophysical Cell Parameters in Digital Holographic Microscopy Images

Authors:

Lilith Brandt, Klaus Brinker and Björn Kemper

Abstract: This paper presents an analysis pipeline for automatically detecting cells in digitally reconstructed quantitative phase images acquired by digital holographic microscopy and for computing biophysical cell parameters. Using an intelligent, integrated image analysis approach, we optimize the overall analysis process which includes several time-consuming, manual steps. The proposed automatic approach shows promising results in an experimental comparison with the current manual evaluation process.

Paper Nr: 66
Title:

Proposal and Validation of a Domaine Specific Language for the Representation of the AGGIR Constants

Authors:

José Manuel Negrete Ramírez, Philippe Roose, Marc Dalmau and Michel Bakni

Abstract: In this paper, we propose a Domain Specific Language (DSL), and we focus on the AGGIR (Autonomie Grontologie Groupes Iso-Ressources) grid model, in order to exploit the variety of interaction capabilities according to the performanceto current activities, users, and context over the time domain. Our aim is providing an analyis tool regarding the coherent behaviour of elderly/handicapped people within a home environment by means of data recovered from sensors using the iCASA simulator. For the DSL validation, we particularly focus on three of the AGGIR variables: dressing, hygiene and transfers; by evaluating their testbility in many scenarios, especially the abnormal ones, which contain readings representing the occurrence of accidents or unexpected behavior of the elderly, in order to see the ability of the system to understand the obtained records correctly and thus generate the appropriate event information.

Paper Nr: 68
Title:

Deep Learning Techniques for Classification of P300 Component

Authors:

Jiří Vaněk and Roman Mouček

Abstract: Deep learning techniques have proved to be beneficial in many scientific disciplines and have beaten stateof- the-art approaches in many applications. The main aim of this article is to improve the success rate of deep learning algorithms, especially stacked autoencoders, when they are used for detection and classification of P300 event-related potential component that reflects brain processes related to stimulus evaluation or categorization. Moreover, the classification results provided by stacked autoencoders are compared with the classification results given by other classification models and classification results provided by combinations of various types of neural network layers.

Paper Nr: 69
Title:

Trust Factors in Healthcare Technology: A Healthcare Professional Perspective

Authors:

Raja Manzar Abbas, Noel Carroll, Ita Richardson and Sarah Beecham

Abstract: Being able to trust technology is of vital importance to its potential users. This is particularly true within the healthcare sector where lives increasingly depend on the correct application of technology to support clinical decision-making. Despite the risk posed by improper use of technology in the healthcare domain, there is a lack of research that examines why healthcare professionals trust healthcare technology. Therefore, there is little evidence regarding the key trust facilitators and barriers. In this paper, we investigate the concept of trust within a healthcare technology context. We conducted a systematic mapping study to identify relevant trust facilitators and barriers in published work in well-known bibliographic databases. Our results present a synthesis of 47 studies that describe trust factors that healthcare professionals associate with healthcare technology. Facilitators include compatibility and perceived systems usefulness, while barriers include privacy concerns and lack of knowledge. We conclude that HCT trust is complex, multi-dimensional, and influenced by a variety of factors at individual and organizational levels.

Paper Nr: 73
Title:

MHealth Technology as a Tool to Promote Blood Donation

Authors:

Joélia Rodrigues da Silva, Christina César Praça Brasil, Bruno Praça Brasil, Larissa Barbosa Paiva, Vinicius Freire de Oliveira, José Eurico de Vasconcelos Filho and Francisco Wandemberg Rodrigues dos Santos

Abstract: The blood donation scenario and its effect on the treatment of patients with hematological diseases and in emergency situations is a constant concern in the health area, requiring guided actions that allow improvements in the donor recruitment and retention processes, and therefore, the increase and maintenance of blood donation. To meet this social demand, an exploratory, qualitative study was carried out with the objective of creating a cell phone application that supports blood donor recruitment and fidelization through prevalent and interactive communication and technological resources, generating social engagement. The new information technologies have been increasingly disseminating in the several social settings as a way of collecting, recording, producing, processing and sharing data and information. A research was carried out to identify the existence of applications that support the activities of the Brazilian blood donation units, integrated to the database and with gamification resources, with none being identified with these characteristics. The proposed tool has differential characteristics in relation to the applications available in the market and may be effective in blood donor recruitment and in supporting health promotion. The application has a deadline of November 2017 to start being used.

Paper Nr: 75
Title:

Unsupervised Temporal Segmentation of Skeletal Motion Data using Joint Distance Representation

Authors:

Christian Lins, Sebastian M. Müller, Max Pfingsthorn, Marco Eichelberg, Alexander Gerka and Andreas Hein

Abstract: In this paper, we present an online method for the unsupervised segmentation of skeletal motion capture data for the assessment of unfavorable or harmful postures in the context of musculoskeletal disorders. The long-time motion capture data is segmented into short motion sequences using joint distances of the captured skeleton. We use the difference between joint distance matrices to detect variances in motion dynamics in which the motion is separated into either a dynamic motion or a static posture. Then, the static posture can be evaluated using well-known posture assessment methods such as the Ovako Working postures Analysing System (OWAS) to derive risk factors for musculoskeletal disorders. The algorithm works in real-time so that it can be incorporated in live warning systems for unfavorable or harmful postures. We evaluated the segmentation algorithm by comparing it with results from state-of-the-art offline motion segmentation algorithms as gold standard. Results show that the algorithm approaches the performance of state-of-the-art offline segmentation algorithms.

Paper Nr: 86
Title:

Design of Health Advice System for Elderly People by Communication Robot

Authors:

Takahito Tamai, Yukie Majima, Hideki Suga, Syuuki Inoue and Kotoka Murashima

Abstract: Recently in Japan, as the declining birth rate and the aging of the population progress, increased medical expenditures and nursing care burdens are presenting great social difficulty. To mitigate that difficulty it is necessary for many elderly people to live independently and healthily. Therefore, this study develops a health advice system to support health promotion for elderly people. We conducted verification experiments to assess the operability and impressions of tablet PCs and robots and to elicit impressions against health advice presented from different devices. Furthermore, we examined the usefulness of the health advice system for elderly people who used the robots.

Paper Nr: 93
Title:

Integrating Person-to-Person Social Support in Smartphone Apps for Promoting Physical Activity

Authors:

Bojan Simoski, Michel Klein, Aart T. van Halteren and Henri Bal

Abstract: The epidemic of physical inactivity is a major health hazard in the modern society, therefore creating effective and innovative health programs and interventions are important. This paper presents a novel approach in which person-to-person social support is incorporated in mHealth interventions for increasing physical activity. Social support is already used as a behaviour change technique in mHealth apps for influencing physical activity, but mostly offered virtually. While virtually-offered social support is efficient, we believe that person-to-person communication, based on personal coaching, might open a new way of influencing the inactive users and their motivation. Responding to this, we developed an Android application that facilitates physically inactive users to connect with a real life coach to receive person-to-person social support. This paper explains the motivation behind the system's design decisions and discusses the potential of social support in mHeatlh apps. In addition, we present the design of the evaluation study in which the hypotheses and research questions will be evaluated.

Paper Nr: 95
Title:

Terminology Enabled Spatio-temporal Analysis and Visualization for Preterm Birth Data in the US

Authors:

Lixia Yao and Kui Wang

Abstract: Preterm birth can lead to many health problems in infants, including brain damage, neurologic disorders, asthma, intestinal problems and vision problems, but the exact cause of preterm birth is unclear. In this study, we investigated if geographic location or the environment can contribute to preterm birth by building a customized data model based on multiple controlled terminologies. We then performed a large-scale quantitative analysis to understand the relationships between the prevalence of preterm birth, the biological mothers’ demographic information and the Metropolitan Statistical Areas (MSAs) of their primary residency from 2010 to 2014. More specifically we considered education, income, race and marital status information of 388 MSAs from the US Census Bureau. The results demonstrated that the overall preterm birth rate for the United States decreased during 2010 to 2014, with Chicago-Naperville-Elgin (Illinois) Metro Area, Houston-Sugar Land (Texas) Metro Area and Billings (Montana) Metro Area observing the most visible improvement. There are statistically significant correlations between race distribution, education level and preterm birth. But median income, marital status and insurance coverage ratio are found irrelevant to preterm birth. This study demonstrated the power of controlled terminologies in integrating medical claims data and geographic data to study preterm birth for first time. The customized common data model and the interactive tool for online visualizing a large preterm dataset from both the temporal and spatial perspectives can be used for future public health studies of many other diseases and conditions.

Paper Nr: 96
Title:

ActiThings Toolkit - Towards Supporting Older Adults’ Adherence to Home-based Physical Exercise Programs by Providing Notifications in Opportune Moments

Authors:

Elke Beck, Kai von Holdt and Jochen Meyer

Abstract: Home-based physical exercise programs can delay or even prevent age-related frailty among older adults, but insufficient adherence to these programs, especially long-term, is a reoccurring problem. We suggest aiding the integration of simple exercises into daily routine by using innovative interactive tools which provide older adults with environmental prompts for exercises in opportune moments. In this paper, we report on our lessons learned with exploring the usefulness of our ActiThings toolkit, which includes a broad range of unobtrusive, portable, ubiquitous sensing and feedback technologies.

Paper Nr: 100
Title:

Applying Process Mining and RTLS for Modeling, and Analyzing Patients’ Pathways

Authors:

Sina Namaki Araghi, Franck Fontanili, Elyes Lamine, Ludovic Tancerel and Frédérick Benaben

Abstract: Purpose: This paper aims at introducing a generic approach for visualizing, analyzing and diagnosing patients’ pathways. This approach could be categorized as a business intelligence approach to extract knowledge for decision makers in healthcare organizations. The analyses provided by this approach are based on the location data which is being recorded in the information system (IS) by indoor-Real-Time Location Systems (RTLS). Findings: Healthcare organizations are getting more eager to learn from the execution of their processes. They seek different tools and approaches to analyze the processes and visualize the problems. This paper presents one of the possibilities to provide more understanding of process executions and it is based on the positions of the patients in the organization. Approach: Business intelligence approaches provide new technical and technological solutions for business analysts to improve the quality of products and services within organizations. The approach in this work helps to visualize patients’ pathways and analyze them by associating real-time localization and process mining. This approach consists of four phases in which several functionalities have been defined. These phases are Data, Information, Awareness, and Governance (DIAG). Also, a case study has been designed to illustrate the DIAG approach.

Paper Nr: 104
Title:

Advances in Building BodyInNumbers Exercise and Wellness Health Strategy Framework

Authors:

Petr Brůha, Roman Mouček, Vítězslav Vacek, Pavel Šnejdar, Lukáš Vařeka, Václav Kraft and Peter Rehor

Abstract: Smoking, excessive drinking, overeating and physical inactivity are well-established risk factors decreasing human physical performance and increasing incidence of chronic diseases. Moreover, epidemiological work has identified modifiable lifestyle factors, such as poor diet, physical and cognitive inactivity that are associated with the risk of reduced cognitive performance. Chronic diseases present an enormous burden to society by increasing medical costs and human suffering. Exercise and wellness health strategy frameworks aiming at influencing modifiable lifestyle risk factors in voluntarily enrolled individuals and thus decreasing incidence of chronic diseases are then very beneficial. However, such frameworks also need a supporting software infrastructure. The advances in building of such software infrastructure, the BodyInNumbers software system for rapid collection and analysis of health related data, are presented in this paper. They include the changes in the system architecture, redefinition of user roles related to data and metadata security and design, implementation and integration of new modules for collection and management of electroencephalographic/P300 event-related potential data and new modules for collection and management of data from measurements of physical strength and balance. The results of the system testing are finally described.

Paper Nr: 108
Title:

Communicating Personalized Risk Factors for Lifestyle Coaching

Authors:

George Drosatos, Kyriakos Bakirlis, Pavlos Efraimidis and Eleni Kaldoudi

Abstract: Chronic non-communicable diseases such as diabetes, chronic cardiorenal and respiratory disease and cancer, are serious, burdensome and costly conditions that share a common characteristic: they heavily depend on common behavioural risk factors, such as physical activity, diet, stress, and substance abuse. Despite concerted efforts it has been remarkably difficult to change such lifestyle related disease determinants, as behavioural change is a complex process requiring significant personal responsibility. In this paper we propose a personal mobile eHealth application to communicate personalized lifestyle related health risks and understand their individual impact on personal health condition and disease progression.

Paper Nr: 115
Title:

Association of Body Mass Index with Estimated Glomerular Filtration Rate and Incident Proteinuria

Authors:

Seung Min Lee, Minseon Park and Hyung-Jin Yoon

Abstract: Obesity has been one of the most important risk factors of chronic kidney disease (CKD). But the association of body mass index (BMI) with estimated glomerular filtration rate (eGFR) and incident proteinuria has not been studied well. The goal of this study was to elucidate the association of BMI with eGFR and proteinuria using nationwide health examination data. These associations were investigated with data of Korean adults who had undergone health screenings at least three times between 2009 and 2014. eGFR was calculated with Chronic Kidney Disease Epidemiology collaboration equation based on serum creatinine level. The association between BMI and eGFR was analysed with a generalized addictive model adjusting for possible confounders. Similarly, the association between BMI and incident proteinuria was analysed with Cox hazard model adjusting for possible confounders. As a result, a V-shape relationship between BMI and eGFR was observed. The nadir was around 29 kg/m2. With subgroup analyses for the association between BMI and eGFR, a V-shape association was observed in men and younger age group and an inverse association was observed in women and older age group. A reverse J-shape association between BMI and the adjusted hazard ratio of incident proteinuria was observed. The nadir was approximately estimated around 22 kg/m2.

Paper Nr: 117
Title:

Developing a Sensor based Homecare System - The Role of Bluetooth Low-Energy in Activity Monitoring

Authors:

Luke Power, Lisa Jackson and Sarah Dunnett

Abstract: Home healthcare systems have become a focus of research due to the shifting care requirements of the elderly. Malnourishment, independence and activity are becoming vital metrics when monitoring patient illness. Monitoring devices described in research however express issues in the consistent remote capture of these metrics. This work presents the role of Bluetooth Low-Energy Beacons (BLE) in community based healthcare by examining how passive activity monitoring can assist patients coping with independence and disease management within their homes as an indoor Proximity System (IPS). BLE sensors will be placed on the patient, in their home and on objects of interest (OOI) such as water bottles, kettles and microwaves. Research described in this paper will focus on accuracy of BLE beacon as an IPS for lifestyle monitoring and its application to intelligent healthcare. This is achieved by creating a model of patient care requirements structured using activities of daily living (ADL) which is evaluated using patient activity pattern recognition in captured sensor data. Pattern analysis uses the changing distance values between BLE sensors to determine movement motion and location which contribute to the activity, sensor based care model. Results support efficacy when using BLE beacons as an IPS with patient activity patterns becoming observable through monitoring with a consistent ability to distinguish interactions in activity patterns capture. Future experiments will focus on analysis captured sensor metrics to determine care outcomes.

Posters
Paper Nr: 2
Title:

Comparison of Decision Tree, Neural Network, Statistic Learning, and k-NN Algorithms in Data Mining of Thyroid Disease Datasets

Authors:

Wafaa Al Somali and Riyad Al Shammari

Abstract: Massive information contained in medical datasets presents challenge to the practitioners in diagnosing diseases or determining health status of patients. Data mining is therefore required to help users obtaining valuable information from a very complex data collection. In this study, we explored several methods of data mining in order to improve the quality of a dataset which is related to diagnosis of thyroid disease. Several classifiers were trained on the dataset and compared to previous study by Akbaş et al (2013). The performance improvement was examined in order to determine the best classifier that can be executed. Findings revealed that decision tree (J48) algorithm outperformed all other algorithms in terms of accuracy, Kappa, Matthew’s correlation coefficient (MCC), and receiver operating characteristics (ROC) with respective values of 0.994, 0.951, 0.953, and 0.987. Classification using J48 was found to be better than those conducted by Akbaş et al. In contrast, IBK algorithm showed the poorest performance, particularly Kappa and MCC. The size of tree generated from J48 and Logistic Model Tree (LMT) varied greatly. Integration of single classifier with AdaBoost classifier mostly resulted in higher accuracy. However, AdaBoost did not improve the performance of NaïveBayes, IBK and RandomForest algorithms. These results were consistent with the previous study using AdaBoost-based ensemble classifier.

Paper Nr: 4
Title:

Using Open Data Kit in a Cluster Randomized Clinical Trial: The Assisted Partner Services Study

Authors:

Paul Macharia, Betsy Sambai, Mathew Dunbar, Beatrice Wamuti, Peter Maingi, Felix Abuna, David Bukusi, Peter Cherutich and Carey Farquhar

Abstract: Paper-based data collection tools are error prone, require more storage space and pose a number of logistical challenges in the collection, analysis and use of study data. Electronic Data Collection (EDC) has the potential to reduce workload, costs, enhances the quality of collected data and is efficacious on study duration. The Assisted Partner Services (APS) a cluster randomized clinical trial (cRCT) based in HIV Testing and Counselling Services (HTS) centres investigated the reliability, accuracy and feasibility of using Open Data Kit (ODK) to collect participant data. Study outcomes showed that interviewers quickly learnt to work with technology. Initial investment in training and pre-testing the EDC tools quickly paid off with increased accuracy and efficiency.

Paper Nr: 13
Title:

An Unsupervised Learning Model for Pattern Recognition in Routinely Collected Healthcare Data

Authors:

Sara Khalid, Andrew Judge and Rafael Pinedo-Villanueva

Abstract: This study examines a large routinely collected healthcare database containing patient-level self-reported outcomes following knee replacement surgery. A model based on unsupervised machine learning methods, including k-means and hierarchical clustering, is proposed to detect patterns of pain experienced by patients and to derive subgroups of patients with different outcomes based on their pain characteristics. Results showed the presence of between two and four different sub-groups of patients based on their pain characteristics. Challenges associated with unsupervised learning using real-world data are described and an approach for evaluating models in the presence of unlabelled data using internal and external cluster evaluation techniques is presented, that can be extended to other unsupervised learning applications within healthcare and beyond. To our knowledge, this is the first study proposing an unsupervised learning model for characterising pain-based patient subgroups using the UK NHS PROMs database.

Paper Nr: 14
Title:

A Design Thinking Approach to Implementing an Android Biometric Unique Identification System for Infant Treatment Follow-up in a Resource Limited Setting

Authors:

Paul Macharia, Nyawira Gitahi-Kamau, Peter Muiruri, Pratap Kumar, Boniface Ngari and Peter Wanganjo

Abstract: Healthcare systems lack individualized follow-up mechanisms for HIV Exposed Infants (HEIs), this unfortunately leads to missed Early Infant Diagnosis (EID) opportunities delaying access to care and treatment. There is a great need for innovative approaches to address the follow-up complexities. A design thinking approach to the use of technology-based interventions offers great out of the box solutions to address healthcare challenges including patient unique identification. Due to its convenience, individuality and efficiency, fingerprint recognition has become the unique identification method of choice. A biometric-based Unique Patient Identifier (UPI) for HEIs would allow their identification and follow-up as they move between health services and facilities improving their treatment outcomes. A secure, reliable and scalable smartphone-based biometric HEI UPI system could improve the provision of HIV Prevention of Mother to Child Transmission (PMTCT) services.

Paper Nr: 18
Title:

Impact of a TV-based Assistive Technology on Older People’s Ability to Self-manage Their Own Health

Authors:

Daniela Loi, Silvia Macis, Danilo Pani, Andrea Ulgheri, Romina Lecis, Marco Guicciardi, Mauro Murgia and Luigi Raffo

Abstract: Nowadays, special emphasis is being focused on involving people on their own health and care. The use of digital technologies in the home-care management process is increasingly contributing to the maintenance of quality of life and preservation of functional independence in older adults. There is a huge number of available m-health applications for self-tracking health parameters, but the majority of them are inconsistent with the needs of older adults who do not currently use technologies such as computers, smartphones or tablets. The aim of this work is to present a pilot study, which included 19 older adults, that was conducted to objectively measure the effect of a TV-based assistive system on the improvement of older adults’ activation levels about self-management of health. The correlation with the usage of specific digital services provided by the system was also investigated. The results reveal how the impact is limited by the aspecific nature of the intervention with respect to the participants’ health condition. At the same time, they are encouraging and indicate that there is the potential for the system to impact on older people’ self-management skills.

Paper Nr: 21
Title:

Predicting 30-day Readmission in Heart Failure using Machine Learning Techniques

Authors:

Jon Kerexeta, Arkaitz Artetxe, Vanessa Escolar, Ainara Lozano and Nekane Larburu

Abstract: Heart Failure (HF) is a syndrome that reduces patients’ quality of life, and has severe impacts on healthcare systems worldwide, such as the high rate of readmissions. In order to reduce the readmissions and improve patients’ quality of life, several studies are trying to assess the risk of a patient to be readmitted, so that taking right actions clinicians can prevent patient deterioration and readmission. Predictive models have the ability to identify patients at high risk. Henceforth, this paper studies predictive models to determine the risk of a HF patient to be readmitted in the next 30 days after discharge. We present two different approaches. In the first one, we combine unsupervised and supervised classification and achieved AUC score of 0.64. In the second one, we combine decision tree and Naïve Bayes classifiers and achieved AUC score of 0.61. Additionally, we discover that the results improve when training the predictive models with different readmission’s threshold outcome, reaching the AUC score of 0.73 when applying the first approach.

Paper Nr: 22
Title:

A Socio-technical Review of Five National Health Information Systems in Ireland using Agreed National Standards

Authors:

Sarah Craig

Abstract: This paper presents a review of five national health information systems in Ireland using a set of guiding principles and information governance standards that have been nationally agreed by the health information regulatory body there. The review uses a socio-technical approach to examine three dimensions of these systems; policy, infrastructure and people. The review was undertaken using documentary analysis of written materials about the systems from both primary and secondary data sources. The findings show that progress has been slow in the development of health information policy in Ireland and as a result, systems like those reviewed vary in how nationally agreed standards and principles have been applied. The paper concludes the need for a more consistent approach to national health information systems like those reviewed using agreed national standards.

Paper Nr: 37
Title:

Speech Technology in Dutch Health Care: A Qualitative Study

Authors:

Ellen Luchies, Marco Spruit and Marjan Askari

Abstract: This study investigates the opportunities of speech technology in Dutch hospitals, and to what extent speech technology can be used for documentation. Furthermore, we clarify why speech technology is used only marginally by Dutch hospital staff. We performed interviews where speech technology users, managers in hospitals and software suppliers were contacted as participants. We then transcribed our interviews and synthesized the pros and cons of speech technology as well as major barriers for the adoption. Our results show various influencing factors that could be clarifications for the fact that only 1% of the medical staff uses speech technology in the Netherlands. The major reasons we found are: speech technology usage at only radiology and pathology departments, \emph{smarttexts} and \emph{smartphrases} of the Electronic Health Record (EHR) compete with speech technology, caregivers have to adjust their way of working which evokes resistance, lack of central authorization at Dutch hospitals and finally, financial barriers. Our results show that speech technology works for radiology and pathology as a tool for documentation, but is found less useful for other departments. For the remaining departments, different applications show potential, such as structured reporting.

Paper Nr: 40
Title:

Lung Cancer Prognosis System using Data Mining Techniques

Authors:

Yomna Omar, Abdullah Tasleem, Michel Pasquier and Assim Sagahyroon

Abstract: This paper describes a Lung Cancer Prognosis System (LCPS) that aims at providing oncologists with an accurate estimate of the health status of their patients. The proposed system is born from two observations: First, lots of efforts are still required in healthcare to improve productivity, accuracy, etc. by providing ad hoc computer-based solutions; second, while increasing popular, AI and data mining tools cannot be used without significant training and expertise. LCPS thus aims at providing the former by integrating the latter into a user-friendly tool, supplementing the knowledge of the expert oncologist with information about their patients, and leading to improved patient care and treatments. LCPS can accept a variety of lung cancer datasets and employs several data mining algorithms to uncover relationships between observed health signs and probable outcomes, and provides oncologists with various statistical results including predictions about their patients’ medical future. Furthermore, LCPS makes it easy to manage patients’ records, allows them view their profiles and any information as deemed suitable by their doctor, including prognosis and other comments. Lastly, while the current application is currently limited to lung cancer treatment, it can be considered a prototype that can be adapted to other diseases.

Paper Nr: 48
Title:

Sharing With Care - Multidisciplinary Teams and Secure Access to Electronic Health Records

Authors:

Mohamed Abomhara, Berglind Smaradottir, Geir M. Køien and Martin Gerdes

Abstract: Ensuring patient privacy and improving patient care quality are two of the most significant challenges faced by healthcare systems around the world. This paper describes the importance and challenges of effective multidisciplinary team treatment and the sharing of patient healthcare records in healthcare delivery. At present, electronic health records are used to create, manage and share patient healthcare information efficiently and effectively. The security and privacy concerns with sharing and the proper use of protected health information need to be highlighted. Additionally, an access control solution is presented, which is suitable for collaborative healthcare systems to address concerns with information sharing and information access. In this access control model, the multidisciplinary team is classified based on Belbin’s team role theory to ensure that access rights are adapted dynamically to the actual needs of healthcare professionals and to guarantee confidentiality as well as protect the privacy of sensitive patient information.

Paper Nr: 55
Title:

Application of an Educative Health Technology in the Training of the Caregiver Family

Authors:

Zélia Maria de Sousa Araújo Santos, Paula Dayanna Sousa dos Santos, Maria Helena de Agrela Gonçalves Jardim, José Manuel Peixoto Caldas, July Grassiely de Oliveira Branco, Mirna Albuquerque Frota, Karla Maria Carneiro Rolim, Amabili Couto Teixeira de Aguiar, Rithianne Frota Carneiro, Ariane Pontes Soares, Ana Carolina Bezerra Moreira and Laurineide de Fátima Diniz Cavalcante

Abstract: The focus of this research is the use of an Educational Technology in Health (ETH) to train the family caregiver in the adherence of hypertensive users to treatment. The participation of the effective family is a factor that can interfere favorably in the other factors, due to its salutary importance in the involvement of the health-disease process of its members. Participant research with the objective of evaluating the changes in the participation of the family caregiver in the adherence of the hypertensive user to the treatment with the application of the Educational Technology in Health, carried out in a Primary Health Care Unit (PHCU) in Fortaleza, Ceará, Brazil, with a group of eleven family caregivers (CF) indicated by the same number of hypertensive users registered in the cited PHCU. The ETH was elaborated based on the assumptions of health education. The ETH consists of 11 (eleven) weekly meetings, with an average duration of sixty minutes. The information was organized through the Bardin content analysis. After the analysis of the results with the application of ETH, we noticed the occurrence of learning among the FC, but in an unequal way. It is seen that the deficit of previous knowledge about arterial hypertension and the treatment differentiated between them, since each one reported diversified experiences.

Paper Nr: 57
Title:

A Computational Pipeline for Sepsis Patients’ Stratification and Diagnosis

Authors:

David Campos, Renato Pinho, Ute Neugebauer, Juergen Popp and José Luis Oliveira

Abstract: Sepsis is still a little acknowledged public health issue, despite its increasing incidence and the growing mortality rate. In addition, a clear diagnosis can be lengthy and complicated, due to highly variable symptoms and non-specific criteria, causing the disease to be diagnosed and treated too late. This paper presents the HemoSpec platform, a decision support system which, by collecting and automatically processing data from several acquisition devices, can help in the early diagnosis of sepsis.

Paper Nr: 72
Title:

Real Time Mortality Risk Prediction: A Convolutional Neural Network Approach

Authors:

Landon Brand, Aditya Patel, Izzatbir Singh and Clayton Brand

Abstract: Machine Learning in Healthcare shows great promise, but is often difficult to implement due to difficulties in collecting data. We used a 1-dimensional convolutional neural network(CNN) on limited data to show a practical application of deep learning in healthcare. We used only vital signs data that can be collected from low cost, readily available hardware designed for non-critical care settings, and a dynamic model that updates as more data is collected over time. Our data is derived from the MIMIC dataset. We use 320 patients for testing and 2,990 for training the model. The CNN model predicted mortalities with up to a 76.3% accuracy, and outperformed both recurrent neural network and multi-layer perceptron models. To our knowledge, the proposed methodology is the first of its kind to predict mortality risk scores based on only heart rate, respiratory rate, and blood pressure, three easily collectible data.

Paper Nr: 89
Title:

Integrative Analysis of CDC and Census Data Revealed Significance of Suicide-related Risk Factor

Authors:

Gilberto Diaz, Jacob Jones, Toni Brandt, Ashwini Yenamandra, Todd Gary and Qingguo Wang

Abstract: Suicide is one of leading causes of death in the United States. By analyzing the data from the Center for Disease Control (CDC) (2003 - 2013), we found the rates of suicide and other manner of deaths of the diseased population who had completed only 12th grade education are considerably higher than the diseased population with different education attainment. By comparing the CDC data with the latest Census data (2010), we also found the high suicide rate of the diseased population that completed only 12th grade education is statically significant. To interrogate suicide in exquisite detail, we also examined data at the state level by incorporating Tennessee suicide data.

Paper Nr: 94
Title:

Perioperative Electronic System - A New Approach for Perioperative Nursing Performance

Authors:

Márcia Baptista, Rita Silva, Helena Gonçalves Jardim and António Quintal

Abstract: Nowadays, IT and informatics are permanently and highly integrated into the delivery of quality health care and in the perioperative care is no exception. At Dr. Nélio Mendonça Hospital the implementation of a perioperative electronic system was a major step. The purpose of this study was to contribute to the perioperative nursing care improvement by recreating innovated nursing practices through the conception and implementation of a perioperative electronic system. Before the perioperative electronic system implementation in the OR only 1,2 % of the nurses registered the preoperative visit and after its implementation 87,6 % of the nurses registered it. The patient features assessed exhibited inferior anxiety levels (1st group: 13,72/ 2nd group: 10,97) and lower pain levels in the preoperative stage (1st group: 2,66/ 2nd group: 1,19), intraoperative stage (1st group: 2,05/ 2nd group: 0,72) and postoperative stage (1st group: 4,5/ 2nd group: 0,45) after the implementation of the perioperative electronic system (p-value <0,05). The results indicate that this system was beneficial to the nurses and to the surgery patients.

Paper Nr: 97
Title:

Stress Dichotomy using Heart Rate and Tweet Sentiment

Authors:

Jaromir Salamon, Kateřina Černá and Roman Mouček

Abstract: Automated detection of human stress from markers is very beneficial for the development of assistive technologies. Blood pressure, skin temperature, galvanic skin response or heart rate are typical physiological markers that help identify human stress. However, not only the human body itself but also the human mood expressed in short text messages can be a useful source of such information about stress. This paper focuses on detection of human stress using two different but synchronized sources of information, human heart rate and sentiment extracted from tweets. During the preliminary experiment lasting for two fifty-day periods, we obtained simultaneously 481 708 heart rate data samples from two wearables and sentiment from 2049 tweets. The tweet data contain a subjective sentiment evaluation that was recorded using positive and negative hashtags. A few states of stress were identified as the result of the data processing. The final discussion provides conclusions and recommendations for future research.

Paper Nr: 98
Title:

Towards a Unified Understanding of eHealth and Related Terms – Proposal of a Consolidated Terminological Basis

Authors:

Lena Otto, Lorenz Harst, Hannes Schlieter, Bastian Wollschlaeger, Peggy Richter and Patrick Timpel

Abstract: The impact of digitization on healthcare gives rise to interdisciplinary concepts such as eHealth. However, achieving improvements in research and innovation requires a valid and unified understanding of the common terminology. Yet, a heterogeneous usage of different terms regarding eHealth can be observed. This leads to a deficient communication between researchers and practitioners, impeding the diffusion, i. e. extensive practical implementation of innovative health concepts. To address this problem, our aim is to consolidate and harmonize eHealth-related terminology. To this end, a literature analysis was conducted to identify established definitions and to formulate a terminological ontology for the related concepts. The current results show a consistent definition of the terms digitization, ICT, and telematics. In contrast, telemedicine, telehealth, eHealth, and mHealth were identified as conflictingly defined terms. Consequently, the proposed ontology serves as a first guidance to support an adequate use of the included terms. Further systematic research of terms is needed to verify the current concept of the ontology. Additionally, specifying the connection between the ontology and the elements of healthcare systems is required for a deeper understanding of the influence of digitization in healthcare.

Paper Nr: 105
Title:

Object Detection Featuring 3D Audio Localization for Microsoft HoloLens - A Deep Learning based Sensor Substitution Approach for the Blind

Authors:

Martin Eckert, Matthias Blex and Christoph M. Friedrich

Abstract: Finding basic objects on a daily basis is a difficult but common task for blind people. This paper demonstrates the implementation of a wearable, deep learning backed, object detection approach in the context of visual impairment or blindness. The prototype aims to substitute the impaired eye of the user and replace it with technical sensors. By scanning its surroundings, the prototype provides a situational overview of objects around the device. Object detection has been implemented using a near real-time, deep learning model named YOLOv2. The model supports the detection of 9000 objects. The prototype can display and read out the name of augmented objects which can be selected by voice commands and used as directional guides for the user, using 3D audio feedback. A distance announcement of a selected object is derived from the HoloLens’s spatial model. The wearable solution offers the opportunity to efficiently locate objects to support orientation without extensive training of the user. Preliminary evaluation covered the detection rate of speech recognition and the response times of the server.

Paper Nr: 107
Title:

Predicting Hospital Capacity and Efficiency

Authors:

James P. McGlothlin, Sriveni Vedire, Hari Srinivasan, Amar Madugula, Srinivasan Rajagopalan and Latifur Khan

Abstract: Hospitals and healthcare systems are challenged to service the growing healthcare needs of the population with limited resources and tightly restrained finances. The best healthcare organizations constantly seek performance improvement by adjusting both resources and processes. However, there are endless options and possibilities for how to invest and adapt, and it is a formidable challenge to choose the right ones. The challenge is that each potential change can have far reaching effects. This challenge is exacerbated even further because it can be very expensive for a hospital to experience logjams in patient movement. Each and every change has a “ripple” effect across the system and traditional analytics cannot calculate all the ramifications and opportunities associated with such changes. This project uses historical records of patient treatment plans in combination with a virtual discrete event simulation model to evaluate and predict capacity and efficiency when resources are added, reduced or reallocated. The model assigns assets as needed to execute the treatment plan, and calculates resulting volumes, length of stay, wait times, cost. This provides a valuable resource to operations management and allows the hospital to invest and allocate resources in ways that maximize financial benefit and quality of patient care.

Paper Nr: 112
Title:

(Re-)Designing the Business Model of a Digital Ecosystem: An Example in the Socio-Care Context

Authors:

Andrea Pistorio, Luca Gastaldi, Paolo Locatelli and Mariano Corso

Abstract: The advent of digital innovations is pushing many companies to re-design their Business Models (BMs). Amir and Zott (2015) described the process concerning the design of a new BM as constituted by elements, themes and antecedents. This research is based on a European project aimed at improving the independent living for elderly people affected by Mild Cognitive Impairment (MCI) or Mild Dementia (MD), through the definition of a new BM based on the adoption of digital innovations. Through a clinical inquiry approach, this research aims at analysing the interactions among antecedents and providing suggestions regarding the tools that could support BM re-design processes for an ecosystem of actors. Results highlighted alternation of antecedents that results in the continuous development of knowledge and increase of collected information. The increasing complexity should be limited thorough the integration of the collected information that allows the removal of not consistent information.

Paper Nr: 116
Title:

Impact of Online Health Information on Patient-physician Relationship and Adherence; Extending Health-belief Model for Online Contexts

Authors:

Tahir Hameed

Abstract: Physicians have information advantage over patients in terms of professional knowledge and expertise, implying patients have to fully depend on them for diagnosis, prescription and treatment. However, in the wake of abundant online health information (OHI) on the internet and through mobile apps, these days patients appear to be better-informed when approaching their physicians. As per health-belief model, patients would be motivated better to adhere to physicians’ prescribed treatments if they feel threatened by their symptoms and/or when they are convinced about the benefits of the treatment. This research proposes improved health-belief model incorporating use of OHI. It identifies different types of OHI shaping up patients’ perceptions prior to interactions with physicians. It suggests that patient-physician meetings (relationship) and consequent adherence behavior of the patients are inter-related and deeply affected by the initial perceptions of the patients based on consumed OHI. The proposed model is being tested using anonymous survey data collected immediately after patient-physician meetings in clinics/hospitals and subsequent adherence data from the same patients. Key contribution of this paper is combining individual’s information behavior with health behavior which provides much better understanding for management of emergent healthcare delivery models in the digital economy.

Paper Nr: 119
Title:

Quality Evaluation of Gamified Blood Donation Apps using ISO/IEC 25010 Standard

Authors:

Ali Idri, Lamyae Sardi and José Luis Fernández-Alemán

Abstract: In the light of the tremendous interest that gamified blood donation (BD) mobile applications (apps) started to gain, it is necessary to consider the quality assessment of the requirements implemented in this apps. In this paper, we provide a general overview of requirements for gamified BD apps, which have been retrieved from literature and extracted from the existing apps on the market. Using the ISO/IEC 25010 quality model, a checklist was established to analyse the influence of the identified requirements on 30 software product quality characteristics. The results obtained show a significant variability in the degree of impact of the various requirements. In particular, users’ actions and App’s features are the blocks of requirements that reached a very high degree of influence on quality characteristics. The only sub-characteristic affected by the whole range of requirements is Appropriateness of Functional Suitability. 92% of requirements influenced Operability characteristic whereas the lowest degrees of impact were noted for Compatibility (16%) and Transferability (11%). Blood donation apps’ developers and stakeholders may consider the degree of impact analysis reported in this study to identify software quality requirements which could be included in the quality assessment of these apps.

Paper Nr: 120
Title:

The Computer-aided Diagnostics of Gastric Lesions by using High Definition Narrow-band Imaging Endoscopy and Real-time Pattern Recognition System

Authors:

K. Yu. Erendzhenova, O. A. Kulagina, R. M. Kadushnikov and T. V. Zarubina

Abstract: High Definition (HD) and Magnified Narrow band imaging endoscopy (ME-NBI) allowed to recognizetypes of gastric lesions according modified VS-classification by professor Yao K., becausethe parameters to describe regular or irregularvascular or microsurface pattern and demarcation line in lesionswere formalized. In this work endoscopic differential criteria of benign and neoplastic epithelial lesions of stomach were obtained. Based on them classification algorithm for the real-time processing of narrow–band endoscopic images with a highly productive distributed intellectual analytic decision support system for multiscale endoscopic diagnostics is presented. We also created the electronic atlas and database to collect high resolution endoscopic images, applied and proved the differential diagnosis of gastric lesions through the computer analysis. The algorithm consistentlyused scale– invariant feature transform detector, computation of gastric mucosa pit–pattern skeletons, “Bag of visual words” method, and K–means method for key pointsclustering. Resulting classification algorithm is completely automated, performed real-time analysis, and did not require preliminary selection of interest area. Image classification accuracy was 85%.