The Karma system is currently undergoing maintenance (Monday, January 29, 2018).
The maintenance period has been extended to 8PM EST.
Karma Credits will not be available for redeeming during maintenance.
The leading peer-reviewed journal for digital medicine, and health & healthcare in the Internet age
The Journal of Medical Internet Research (JMIR), now in its' 20th year, is the pioneering open access eHealth journal, and is the flagship journal of JMIR Publications. It is the leading digital health journal, in terms of quality/visibility (Impact Factor 2016: 5.175, ranked #1 out of 22 journals) and in terms of size (number of papers published). The journal focuses on emerging technologies, medical devices, apps, engineering, and informatics applications for patient education, prevention, population health and clinical care. As leading high-impact journal in its' disciplines (health informatics and health services research), it is selective, but it is now joined by almost 30 specialty JMIR sister journals, which have a broader scope (peer-review reports are portable across JMIR journals).
As open access journal we are read by clinicians and patients alike and have (as all JMIR journals) a focus on readable and applied science reporting the design and evaluation of health innovations and emerging technologies. We publish original research, viewpoints, and reviews (both literature reviews and medical device/technology/app reviews).
Be a widely cited leader in the digitial health revolution and submit your paper today!
Right click to copy or hit: ctrl+c (cmd+c on mac)
Background: The clinical heterogeneity of patients with diabetes increases the challenge of maintaining glycemic control and therapy adherence. It is fundamental that patients are actively involved in...
Background: The clinical heterogeneity of patients with diabetes increases the challenge of maintaining glycemic control and therapy adherence. It is fundamental that patients are actively involved in the management of the disease in their living environments. This requires taking medicines, following a proper diet, trying to do some exercise, and being educated about and trained on the condition and its main risks. Objective: To build on top of User Centered Design techniques a supporting self-management system for diabetes in community settings. The aim is to show how User Centered Design techniques can be relevant to define and personalize eHealth solutions for the management of diabetes by assessing the use and compliance of the self-management system in a small-scale multicenter randomized study Methods: User Centered Design principles was used to involve diabetic patients and treating rpofessionals into the design, development and evaluation of a self-management system. An adaptation of the G-OD methodology was used throughout the whole process, through three main iterative cycles: scenario definition, user archetype definition and system development. Usage and compliance metrics were defined for assessing the level of engagement of patients towards their prescribed treatment and the adherence to the proposed self-management system during a four week duration study. The Wilcoxon test was chosen as a particularly conservative method, sacrificing test-power for accurateness under possibly non-parametric conditions for comparing the observed measurements for each of the weeks in the study. Results: A comprehensive system incorporating modules for the management of blood glucose levels, medication, food intake habits, physical activity, diabetic education and messaging was adapted to each of the two type of principal users (personas): Type 1 Diabetes user and Type 2 Diabetes User. 20 patients and 24 treating professionals enrolled the study and used the implemented system for a period of four weeks. The assessment of usage and compliance metrics was similar and did not achieve a significant difference among the two type of users, except to the medication module, which show a significant different use and compliance (P=.01) Conclusions: A self-management system for diabetes based on User Centered Design empowered patients to make their own decisions and help to expand the concept of Personal Health Records as an addition to the Electronic Health Records. After an initial period of time, T1DM are more prone to use less the parts of the designed app, however the use and the communications sent from the app still remain similar among T1DM and T2DM patients
Background: Social Network Sites (SNSs) are increasingly being used to exchange health information from patients and practitioners/pharmaceutical companies/research centers. Research contributions hav...
Background: Social Network Sites (SNSs) are increasingly being used to exchange health information from patients and practitioners/pharmaceutical companies/research centers. Research contributions have explored the contents discussed online, categorized the topics, and explored their engagement levels. Objective: This research aims at investigating the potential role of Social Networks Site (SNSs) in Healthcare. Specifically it provides a clustering of health information available on SNSs and creates an initial research design that would allow the use of this information to enhance healthcare delivery. In addition, this research aims at testing whether SNSs valid tools for sharing drug related information by patients. Methods: The research is based on a specific chronic disease: Multiple Sclerosis. We searched the SNS Facebook and looked at all existing groups on this condition. The analysis was restricted to public groups for privacy concerns. We created a database by downloading posts from two main groups on which we performed a content analysis and a statistical analysis; this allowed us to discriminate between categories, their engagement level, and type of posts shared. The mean of engagement for each topic was analyzed using one-way ANOVA and followed up by pairwise comparisons using TukeyHSD. Results: On a sample of 7029 posts, initial results show that there are 8 categories of topics that have resonance (percentage of times the topic appears in our sample) with those who post on Facebook: Patient Support (3.09%), Information/Awareness (70.02%), Event Advertising and Petitions (5.19%), Products and Drugs Advertising (0.68%), Fundraising (5.04%), Clinical Trials or Research Studies (0.84%), Drug Discussion (2.05%), and Other (14.14%). Initial analysis shows that “comments” and “likes” (as measures of engagement level) are more frequently used than other measures of engagement. The results show high engagement level (in terms of views, likes, comments, etc.) for Patient support, Information/Awareness. In addition, although Drug Discussion had low resonance it had unexpected highly engagement level which we found worthy of further exploration. Conclusions: SNSs have become important tools for patients and healthcare practitioners to share or seek information. We identify the type of information shared and how the public reacts to it. Our research confirms that the categories of topics discussed in social media related to specific diseases are appropriate as they are similar to the categories observed by other researchers. Additionally, we found other categories such as drug discussion which was unexpected. This and other results of our study enhance our understanding of how contents are disseminated and perceived within a specific disease based community. We conclude that this information has useful implications in the design of prevention campaigns, educational programs, and chronic disease management.
Background: For advanced cancer patients in palliative care, a crucial phase is the transition from palliative care in the hospital to the home setting, where 24-7 care is not guaranteed any more. To...
Background: For advanced cancer patients in palliative care, a crucial phase is the transition from palliative care in the hospital to the home setting, where 24-7 care is not guaranteed any more. To fill this gap after transition, we are evaluating the feasibility of a physical and social activity tracking system consisting of a FDA approved bracelet (Biovotion Everion MD®) collecting vital data, e.g., heart rate, oxygen saturation etc., and an Android smart-phone (Samsung Galaxy S5) collecting patients’ self-reports of pain and distress as well as acceleration, GPS and phone call statistics data. When study participants are asked, how they are doing in general, a common answer is “There are good days and there are bad days.” Apparently, they order their days into different groups. We argue that these “good” and “bad” days have impact on a patient’s behavior and is therefore visible in the collected activity data. Objective: As a part of the study’s goals, we aim to show the explanatory power of the collected data: the collected data reflect the health status of a patient. Methods: Data is collected over a study period of 12 weeks as part of a feasibility study with an explorative and descriptive study design. Study participants are enrolled from the wards of the Clinic of Radiation-Oncology at the University Hospital Zurich, including the specialized palliative care ward. The data collection chain consists of the patients’ devices, Wi-Fi and internet for secured data upload and a receiving web server. The raw data is preprocessed involving resampling and basic feature extraction. Complex features are extracted using unsupervised machine learning methods, e.g., clustering. Heat maps are used to provide overview visualizations of sensor modalities. Integrated views are generated for multi-modal reconstruction and visualization of patients’ daily routines. Results: Data collection started in March 2017 and already 13 study participants have finished their study participation or had to abort their participation due to health reasons. We collected more than 10000 hours of valid bracelet data and about 410000 GPS positions from the smart-phone. The cohort shows a high variability in live circumstances, e.g., some are still working, and others hardly leave their homes. We give examples of two patients with different courses of disease in order to demonstrate our approach. Conclusions: Our remote monitoring system delivers a large amount of data that allows us to reconstruct the daily routines of the patients showing differences between good and bad days. Clinical Trial: The local Ethics Committee (Kantonale Ethikkommission Zürich) has approved the study protocol; approval number PB_2016-00895.
Background: Background: Sports related concussion forms a major component of all brain injuries occurring in the United States and has a huge detrimental impact on the quality of life and various heal...
Background: Background: Sports related concussion forms a major component of all brain injuries occurring in the United States and has a huge detrimental impact on the quality of life and various health outcome. Predicting concussion is an important way to achieve prevention. Understanding concussion likelihood in the context of different data such as demographic, life style and mental health information related injury will support the development of better diagnostics and preventative techniques. Objective: The objective of this study is to predict the concussion occurrence, number of the concussion, and number of the years since the last concussion using the analytical models. Methods: We develop analytic models that are built using disparate data about lifestyle, demographics and medical history. These models that are based on various machine learning algorithms such as K_Nearest Neighbor, Support Vector Machines, Regression, Ensemble models, Artificial Neural Networks, Decision Tree, General Linear Model and Multivariate Adaptive Regression Splines. In this paper the synthetic minority over-sampling (SMOTE) is employed to overcome the data-imbalance problems. Results: The results show that the predictors associated with the cognitive-mental health plays an important role as a predictor of concussions. Findings suggest that Random forest, Artificial Neural Networks and Decision Tree demonstrate superior performance (sensitivity-80, specificity-88, accuracy- 86) over the other analytics approaches. The number of the concussions are best predicted by K_Nearest Neighbor (sensitivity-83, specificity-75, accuracy-80) while Multivariate Adaptive Regression Splines (mean absoluter error - 2.45) and General Linear Model (mean absoluter error - 2.67) outperform the other machine learning methods for predicting the number of the years past from last concussion Conclusions: Using the data derived from a series of easily executable screening test supported with IoT devices and self-reports, comprehensive analytics models to predict concussion occurrence, reoccurrence and duration since last concussion based on their demographic, lifestyle and mental health information can be developed. Such computational models could lead to customized training approaches and improved efforts for concussion prevention and management.
Background: A smartphone is a promising tool for daily cardiovascular measurement and mental stress monitoring. Photoplethysmography (PPG) and low-cost thermography can be used to create cheap, conven...
Background: A smartphone is a promising tool for daily cardiovascular measurement and mental stress monitoring. Photoplethysmography (PPG) and low-cost thermography can be used to create cheap, convenient and mobile systems. However, to achieve robustness, a person has to remain still for several minutes while a measurement is being taken. This is very cumbersome, and limits the usage in applications such producing instant measurements of stress. Objective: We propose to use smartphone-based mobile PPG and thermal imaging to provide a fast binary measure of stress responses to an event using dynamical physiological changes which occur within 20 seconds of the event finishing. Methods: We propose a system that uses a smartphone and its physiological sensors to reliably and continuously measure over a short window of time a person’s blood volume pulse, the time interval between heartbeats (R-R interval) and the 1D thermal signature of the nose tip. 17 healthy participants, involved in a series of stress-inducing mental activities, measured their physiological response to stress in the 20 second-window immediately following each activity. A 10-cm Visual Analogue Scale was used by them to self-report their level of mental stress. As a main labeling strategy, normalized K-means clustering is used to better treat inter-personal differences in ratings. By taking an array of the R-R intervals and thermal directionality as a low-level feature input, we mainly use an artificial neural network to enable the automatic feature learning and the machine learning inference process. To compare the automated inference performance, we also extracted widely used high level features from HRV (e.g., LF/HF ratio) and the thermal signature and input them to a k-nearest neighbor to infer perceived stress levels. Results: First, we tested the physiological measurement reliability. The measured cardiac signals were considered highly reliable (signal goodness probability used, Mean=0.9584, SD=0.0151). The proposed 1D thermal signal processing algorithm effectively minimized the effect of respiratory cycles on detecting the apparent temperature of the nose tip (respiratory signal goodness probability Mean=0.8998 to Mean=0). Second, we tested the 20 seconds instant perceived stress inference performance. The best results were obtained by using automatic feature learning and classification using artificial neural networks rather than using pre-crafted features. The combination of both modalities produced higher accuracy on the binary classification task using 17-fold leave-one-subject-out (LOSO) cross-validation (accuracy: HRV+Thermal: 76.96%; HRV: 60.29%; Thermal: 61.37%). The results are comparable with the state of the art automatic stress recognition methods requiring long term measurements (a minimum of 2 minutes for up to around 80% accuracy from LOSO). Lastly, we explored the impact of different data labeling strategies used in the field on the sensitivity of our inference methods and the need for normalization within individual. Conclusions: Results demonstrate the capability of smartphone biomedical imaging in instant mental stress recognition. Given that this approach does not require long measurements requiring attention and reduced mobility, it is more feasible for mobile mental healthcare solution in the wild.
Background: Ambient Assisted Living (AAL) solutions have been conquering an important place among strategies to promote ageing in place and address the societal challenges of population ageing. AAL is...
Background: Ambient Assisted Living (AAL) solutions have been conquering an important place among strategies to promote ageing in place and address the societal challenges of population ageing. AAL is deeply rooted on the computing paradigm of Ambient Intelligence which strongly impacts the technological phenomenon of Internet of Things (IoT), currently covering a plethora of ageing related application areas. The pervasiveness of IoT raise, however, security challenges and require more flexible and better adapted availability and privacy measures. Still, IoT devices and services are frequently described in the literature without any reference to privacy and security issues they may integrate and the few works in the Ambient Assisted Living (AAL) field focus mostly on authentication or physical access control. Objective: This paper describes the SoTRAACE - Socio-Technical Risk-Adaptable Access Control - model, designed to better adapt users’ access control needs to each AAL security context. The model is applied to use cases based on AAL for mental health personas and scenarios. Methods: SoTRAACE architecture takes into account contextual, technological and user’s interaction profiling functionalities to act in each AAL situation/request and perform a quantitative and qualitative risk assessment analysis. The risk analysis supports decision-making on the most secure, private and usable way to access and display information. Results: SoTRAACE unique advantages for improved availability and privacy are discussed in contrast with existing access control models. The model is showcased and discussed within two AAL for mental health use case scenarios. SoTRAACE new and reused components are varied and versatile enough to adapt to different situations and user’s goals, whether these are patient or caregiver oriented. Conclusions: SoTRAACE is an innovative and complete proposal for secure and adaptable access control in AAL or similar environments.