Models and Data Sources Used in Systems Medicine. A Systematic Literature Review.

BACKGROUND: Systems medicine is a new approach for the development and selection of treatment strategies for patients with complex diseases. It is often referred to as the application of systems biology methods for decision making in patient care. For systems medicine computer applications, many different data sources have to be integrated and included into models. This is a challenging task for Medical Informatics since the approach exceeds traditional systems like Electronic Health Records. To prioritize research activities for systems medicine applications, it is necessary to get an overview over modelling methods and data sources already used in this field. OBJECTIVES: We performed a systematic literature review with the objective to capture current use of 1) modelling methods and 2) data sources in systems medicine related research projects. METHODS: We queried the MEDLINE and ScienceDirect databases for papers associated with the search term systems medicine and related terms. Papers were screened and assessed in full text in a two-step process according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement guidelines. RESULTS: The queries returned 698 articles of which 34 papers were finally included into the study. A multitude of modelling approaches such as machine learning and network analysis was identified and classified. Since these approaches are also used in other domains, no methods specific for systems medicine could be identified. Omics data are the most widely used data types followed by clinical data. Most studies only include a rather limited number of data sources. CONCLUSIONS: Currently, many different modelling approaches are used in systems medicine. Thus, highly flexible modular solutions are necessary for systems medicine clinical applications. However, the number of data sources included into the models is limited and most projects currently focus on prognosis. To leverage the potential of systems medicine further, it will be necessary to focus on treatment strategies for patients and consider a broader range of data.

Multimodal sensor-based fall detection within the domestic environment of elderly people.

BACKGROUND: Falls represent a major threat to the health of the elderly and are a growing burden on the healthcare systems. With the growth of the elderly population within most societies efficient fall detection becomes increasingly important; however, existing fall detection systems still fail to produce reliable results. OBJECTIVES: A study was carried out on sensor-based fall detection, analysis of falls with the help of fall protocols and the analysis of user acceptance of fall detection sensor technology through questionnaires. MATERIAL AND METHODS: A total of 28 senior citizens were recruited from a German community-dwelling population. The primary goal was a sensor-based detection of falls with accelerometers, video cameras and microphones. Details of the falls were analyzed with the help of medical geriatric assessments and standardized fall protocols. The study duration was 8 weeks and required a maximum of nine visits per subject. RESULTS: The study participants were 28 subjects with a mean age of 74.3 and a standard deviation (SD) of ± 6.3 years of which 12 were male and 16 female. A total of 1225.7 measurement days were recorded from all participants and the algorithms detected 2.66 falls per day. During the study period 15 falls occurred and 12 of these falls were correctly recognized by the fall detection system. CONCLUSION: Current fall detection technologies work well under laboratory conditions but it is still problematic to produce reliable results when these technologies are applied to real life conditions. Acceptance towards the sensors decreased after study participation although the system was generally perceived as useful or very useful.

Home-centered health-enabling technologies and regional health information systems. An integration approach based on international standards.

INTRODUCTION: This article is part of the Focus Theme of Methods of Information in Medicine on "Using Data from Ambient Assisted Living and Smart Homes in Electronic Health Records". OBJECTIVES: In this paper, we present a prototype of a Home-Centered Health-Enabling Technology (HET-HC), which is able to capture, store, merge and process data from various sensor systems at people's home. In addition, we present an architecture designed to integrate HET-HC into an exemplary regional Health Information System (rHIS). METHODS: rHIS are traditionally document-based to fit to the needs in a clinical context. However, HET-HC are producing continuous data streams for which documents might be an inappropriate representation. Therefore, the HET-HC could register placeholder-documents at rHIS. These placeholder-documents are assembled upon user-authenticated request by the HET-HC and are always up-to-date. Moreover, it is not trivial to find a clinical coding system for continuous sensor data and to make the data machine-readable in order to enhance the interoperability of such systems. Therefore, we propose the use of SNOCAP-HET, which is a nomenclature to describe the context of sensor-based measurements in health-enabling technologies. RESULTS: We present an architectural approach to integrate HET-HC into rHIS. Our solution is the centralized registration of placeholder-documents with rHIS and the decentralized data storage at people's home. CONCLUSIONS: We concluded that the presented architecture of integrating HET-HC into rHIS might fit well to the traditional approach of document-based data storage. Data security and privacy issues are also duly considered.

Measurement of accelerometry-based gait parameters in people with and without dementia in the field: a technical feasibility study.

BACKGROUND: Gait analyses are an important tool to diagnose diseases or to measure the rehabilitation process of patients. In this context, sensor-based systems, and especially accelerometers, gain in importance. They are able to improve objectiveness of gait analyses. In clinical settings, there is usually a supervisor who gives instructions to the patients, but this can have an influence on patients' gait. It is expected that this effect will be smaller in field studies. OBJECTIVE: Aim of this study was to capture and evaluate gait parameters measured by a single waist-mounted accelerometer during everyday life of subjects. METHODS: Due to missing ground-truth in unsupervised conditions, another external criterion had to be chosen. Subjects of two different groups were considered: patients with dementia (DEM) and active older people (ACT). These groups were chosen, because of the expected difference in gait. The idea was to quantify the expected difference of accelerometric-based gait parameters. Gait parameters were e.g. velocity, step frequency, compensation movements, and variance of the accelerometric signal. RESULTS: Ten subjects were measured in each group. The number of walking episodes captured was 1,187 (DEM) vs. 1,809 (ACT). The compensation and variance parameters showed an AUC value (Area Under the Curve) between 0.88 and 0.92. In contrast, velocity and step frequency performed poorly (AUC values of 0.51 and 0.55). It was possible to classify both groups using these parameters with an accuracy of 89.2%. CONCLUSION: The results showed a much higher amount of walking episodes in field studies compared to supervised clinical trials. The classification showed a high accuracy in distinguishing between both groups.

GAL@Home: a feasibility study of sensor-based in-home fall detection.

BACKGROUND: A considerable proportion of falls occur within the domestic environment. Sensor-based identification of falls in seniors' homes could help them to remain autonomous and self-sufficient in their own homes. The objective of this study was to evaluate fall detection systems within the home environment using optical and accelerometric sensor systems. METHODS: Portable triaxial accelerometers and optical sensors were used to detect falls in subjects with known problems of mobility and a recent fall history. RESULTS: Three subjects were investigated with the system. Overall nine falls occurred during the study period. Four falls were recorded by the accelerometric system and one fall by the optical system. Subjects with increased risk of falling as measured with mobility and fall risk assessments tend to fall more frequently. CONCLUSION: The study shows that there is a considerably large difference between fall-detector evaluation studies in domestic environments and in laboratory trials.

Development and clinical validation of an unobtrusive ambulatory knee function monitoring system with inertial 9DoF sensors.

Patients suffering from end-stage knee osteoarthritis are often treated with total knee arthroplasty, improving their functional mobility. A number of patients, however, report continued difficulty with stair ascent and descent or sportive activity after surgery and are not completely satisfied with the outcome. State-of-the-art analyses to evaluate the outcome and mobility after knee replacement are conducted under supervised settings in specialized gait labs and thus can only reflect a short period of time. A number of external factors may lead to artificial gait patterns in patients. Moreover, clinically relevant situations are difficult to simulate in a stationary gait lab. In contrast to this, inertial sensors may be used additionally for unobtrusive gait monitoring. However, recent notable approaches found in literature concerning knee function analysis have so far not been applied in a clinical context and have therefore not yet been validated in a clinical setting. The aim of this paper is to present a system for unsupervised long-term monitoring of human gait with a focus on knee joint function, which is applicable in patients' everyday lives and to report on the validation of this system gathered during walking with reference to state-of-the-art gait lab data using a vision system (VICON Motion System). The system KINEMATICWEAR - developed in close collaboration of computer scientists and physicians performing knee arthroplasty - consists of two sensor nodes with combined tri-axial accelerometer, gyroscope and magnetometer to be worn under normal trousers. Reliability of the system is shown in the results. An overall correlation of 0.99 (with an overall RMSE of 2.72) compared to the state-of-the-art reference system indicates a sound quality and a high degree of correspondence. KINEMATICWEAR enables ambulatory, unconstrained measurements of knee function outside a supervised lab inspection.

Sensor-based fall risk assessment--an expert 'to go'.

BACKGROUND: Falls are a predominant problem in our aging society, often leading to severe somatic and psychological consequences, and having an incidence of about 30% in the group of persons aged 65 years or above. In order to identify persons at risk, many assessment tools and tests have been developed, but most of these have to be conducted in a supervised setting and are dependent on an expert rater. OBJECTIVES: The overall aim of our research work is to develop an objective and unobtrusive method to determine individual fall risk based on the use of motion sensor data. The aims of our work for this paper are to derive a fall risk model based on sensor data that may potentially be measured during typical activities of daily life (aim #1), and to evaluate the resulting model with data from a one-year follow-up study (aim #2). METHODS: A sample of n = 119 geriatric inpatients wore an accelerometer on the waist during a Timed 'Up & Go' test and a 20 m walk. Fifty patients were included in a one-year follow-up study, assessing fall events and scoring average physical activity at home in telephone interviews. The sensor data were processed to extract gait and dynamic balance parameters, from which four fall risk models--two classification trees and two logistic regression models--were computed: models CT#1 and SL#1 using accelerometer data only, models CT#2 and SL#2 including the physical activity score. The risk models were evaluated in a ten-times tenfold cross-validation procedure, calculating sensitivity (SENS), specificity (SPEC), positive and negative predictive values (PPV, NPV), classification accuracy, area under the curve (AUC) and the Brier score. RESULTS: Both classification trees show a fair to good performance (models CT#1/CT#2): SENS 74%/58%, SPEC 96%/82%, PPV 92%/ 74%, NPV 77%/82%, accuracy 80%/78%, AUC 0.83/0.87 and Brier scores 0.14/0.14. The logistic regression models (SL#1/SL#2) perform worse: SENS 42%/58%, SPEC 82%/ 78%, PPV 62%/65%, NPV 67%/72%, accuracy 65%/70%, AUC 0.65/0.72 and Brier scores 0.23/0.21. CONCLUSIONS: Our results suggest that accelerometer data may be used to predict falls in an unsupervised setting. Furthermore, the parameters used for prediction are measurable with an unobtrusive sensor device during normal activities of daily living. These promising results have to be validated in a larger, long-term prospective trial.

Defining the user requirements for wearable and optical fall prediction and fall detection devices for home use.

One of the major problems in the development of information and communication technologies for older adults is user acceptance. Here we describe the results of focus group discussions that were conducted with older adults and their relatives to guide the development of assistive devices for fall detection and fall prevention. The aim was to determine the ergonomic and functional requirements of such devices and to include these requirements in a user-centered development process. A semi-structured interview format based on an interview guide was used to conduct three focus group discussions with 22 participants. The average age was 75 years in the first group, 68 years in the second group and 50 years in the third group (relatives). Overall, participants considered a fall prediction system to be as important as a fall detection system. Although the ambient, unobtrusive character of the optical sensor system was appreciated, wearable inertial sensors were preferred because of their wide range of use, which provides higher levels of security. Security and mobility were two major reasons for people at risk of falling to buy a wearable and/or optical fall prediction and fall detection device. Design specifications should include a wearable, non-stigmatising sensor at the user's wrist, with an emergency option in case of falling.

Decision support for teletraining of COPD patients.

BACKGROUND: Supervised physical training has been shown to promote rehabilitation of patients affected by chronic obstructive pulmonary disease (COPD). Currently, due to limited resources, not all COPD patients can be trained by an expert supervisor. OBJECTIVES: The objective of our research is to construct a decision support system (DSS) which observes and controls physical ergometer training sessions of COPD patients. METHODS: A systematic literature review and expert interviews were carried out to build up the knowledge base for the DSS. RESULTS: Nine production rules were established and standardized by Drools and Arden Syntax. The developed software autonomously controls training sessions on a bicycle ergometer on the basis of vital signs data. Thus it offers a new way for the rehabilitation of COPD patients. CONCLUSION: Evaluation with nine healthy subjects in a laboratory environment has confirmed its correct function, but the effects of its use for COPD patients' rehabilitation and their quality of life have to be investigated in a further study.

Predicting in-patient falls in a geriatric clinic: a clinical study combining assessment data and simple sensory gait measurements.

BACKGROUND: Falls are among the predominant causes for morbidity and mortality in elderly persons and occur most often in geriatric clinics. Despite several studies that have identified parameters associated with elderly patients' fall risk, prediction models -- e.g., based on geriatric assessment data -- are currently not used on a regular basis. Furthermore, technical aids to objectively assess mobility-associated parameters are currently not used. OBJECTIVES: To assess group differences in clinical as well as common geriatric assessment data and sensory gait measurements between fallers and non-fallers in a geriatric sample, and to derive and compare two prediction models based on assessment data alone (model #1) and added sensory measurement data (model #2). METHODS: For a sample of n=110 geriatric in-patients (81 women, 29 men) the following fall risk-associated assessments were performed: Timed 'Up & Go' (TUG) test, STRATIFY score and Barthel index. During the TUG test the subjects wore a triaxial accelerometer, and sensory gait parameters were extracted from the data recorded. Group differences between fallers (n=26) and non-fallers (n=84) were compared using Student's t-test. Two classification tree prediction models were computed and compared. RESULTS: Significant differences between the two groups were found for the following parameters: time to complete the TUG test, transfer item (Barthel), recent falls (STRATIFY), pelvic sway while walking and step length. Prediction model #1 (using common assessment data only) showed a sensitivity of 38.5% and a specificity of 97.6%, prediction model #2 (assessment data plus sensory gait parameters) performed with 57.7% and 100%, respectively. DISCUSSION AND CONCLUSION: Significant differences between fallers and non-fallers among geriatric in-patients can be detected for several assessment subscores as well as parameters recorded by simple accelerometric measurements during a common mobility test. Existing geriatric assessment data may be used for falls prediction on a regular basis. Adding sensory data improves the specificity of our test markedly.