Predicting stroke outcome: A case for multimodal deep learning methods with tabular and CT Perfusion data.

MOTIVATION: Acute ischemic stroke is one of the leading causes of morbidity and disability worldwide, often followed by a long rehabilitation period. To improve and personalize stroke rehabilitation, it is essential to provide a reliable prognosis to caregivers and patients. Deep learning techniques might improve the predictions by incorporating different data modalities. We present a multimodal approach to predict the functional status of acute ischemic stroke patients after their discharge based on tabular data and CT perfusion imaging. METHODS: We conducted experiments on tabular, imaging, and multimodal deep learning architectures to predict dichotomized mRS scores 3 months after the event. The dataset was collected from a Dutch hospital and includes 98 CVA patients with a visible occlusion on their CT perfusion scan. Tabular data is based on the Dutch Acute Stroke Audit data, and imaging data consists of summed-up CT perfusion maps. RESULTS: On the tabular data, TabNet outperformed our baselines with an AUC of 0.71, while ResNet-10 on the imaging data performed comparably with an AUC of 0.70. Our implementation of the multimodal DAFT architecture outperforms baselines as well as comparable studies by achieving an 0.75 AUC, and 0.80 F1 score. This was achieved with a final model of less than a hundred thousand optimizable parameters, and a dataset less than half the size of reference papers. CONCLUSION: Overall, we demonstrate the feasibility of predicting the functional outcome for ischemic stroke patients and the usability of multimodal deep learning architectures for this purpose.

Effectiveness of a Pharmacist-Led Web-Based Medication Adherence Tool With Patient-Centered Communication: Results of a Clustered Randomized Controlled Trial.

BACKGROUND: Growing numbers of people use medication for chronic conditions; nonadherence is common, leading to poor disease control. A web-based tool to identify an increased risk for nonadherence with related potential individual barriers might facilitate tailored interventions and improve adherence. OBJECTIVE: This study aims to assess the effectiveness of a newly developed tool aimed at improving medication adherence. METHODS: We performed a cluster randomized controlled trial in patients initiating cardiovascular or oral blood glucose-lowering medication. Participants were recruited from community pharmacies. They completed an online questionnaire comprising assessments of their risk for medication nonadherence and subsequently of barriers to adherence. In pharmacies belonging to the intervention group, individual barriers displayed in a graphical profile on a tablet were discussed by pharmacists and patients with high nonadherence risk in face-to-face meetings and shared with their general practitioners and practice nurses. Tailored interventions were initiated by pharmacists. Barriers of control patients were not presented nor discussed and these patients received usual care. The primary outcome was the effectiveness of the intervention on medication adherence at 8 months' follow-up between patients with an increased nonadherence risk from the intervention and control groups, calculated from dispensing data. RESULTS: Data from 492 participants in 15 community pharmacies were available for analyses (intervention 253, 7 pharmacies; control 239, 8 pharmacies). The intervention had no effect on medication adherence (B=-0.01; 95% CI -0.59 to 0.57; P=.96), nor in the post hoc per-protocol analysis (B=0.19; 95% CI -0.50 to 0.89; P=.58). CONCLUSIONS: This study showed no effectiveness of a risk stratification and tailored intervention addressing personal barriers for medication adherence. Various potential explanations for lack of effectiveness were identified. These explanations relate, for instance, to high medication adherence in the control group, study power, and fidelity. Process evaluation should elicit possible improvements and inform the redesign of intervention and implementation. TRIAL REGISTRATION: The Netherlands National Trial Register NTR5186; https://tinyurl.com/5d8w99hk.

Understanding the complexities of Bluetooth for representing real-life social networks: A methodology for inferring and validating Bluetooth-based social network graphs.

Bluetooth (BT) data has been extensively used for recognizing social patterns and inferring social networks, as BT is widely present in everyday technological devices. However, even though collecting BT data is subject to random noise and may result in substantial measurement errors, there is an absence of rigorous procedures for validating the quality of the inferred BT social networks. This paper presents a methodology for inferring and validating BT-based social networks based on parameter optimization algorithm and social network analysis (SNA). The algorithm performs edge inference in a brute-force search over a given BT data set, for deriving optimal BT social networks by validating them with predefined ground truth (GT) networks. The algorithm seeks to optimize a set of parameters, predefined considering some reliability challenges associated to the BT technology itself. The outcomes show that optimizing the parameters can reduce the number of BT data false positives or generate BT networks with the minimum amount of BT data observations. The subsequent SNA shows that the inferred BT social networks are unable to reproduce some network characteristics present in the corresponding GT networks. Finally, the generalizability of the proposed methodology is demonstrated by applying the algorithm on external BT data sets, while obtaining comparable results.

Digital Health Behavior Change Technology: Bibliometric and Scoping Review of Two Decades of Research.

BACKGROUND: Research on digital technology to change health behavior has increased enormously in recent decades. Due to the interdisciplinary nature of this topic, knowledge and technologies from different research areas are required. Up to now, it is not clear how the knowledge from those fields is combined in actual applications. A comprehensive analysis that systematically maps and explores the use of knowledge within this emerging interdisciplinary field is required. OBJECTIVE: This study aims to provide an overview of the research area around the design and development of digital technologies for health behavior change and to explore trends and patterns. METHODS: A bibliometric analysis is used to provide an overview of the field, and a scoping review is presented to identify the trends and possible gaps. The study is based on the publications related to persuasive technologies and health behavior change in the last 18 years, as indexed by the Web of Science and Scopus (317 and 314 articles, respectively). In the first part, regional and time-based publishing trends; research fields and keyword co-occurrence networks; influential journals; and collaboration network between influential authors, countries, and institutions are examined. In the second part, the behavioral domains, technological means and theoretical foundations are investigated via a scoping review. RESULTS: The literature reviewed shows a clear and emerging trend after 2001 in technology-based behavior change, which grew exponentially after the introduction of the smartphone around 2009. Authors from the United States, Europe, and Australia have the highest number of publications in the field. The three most active research areas are computer science, public and occupational health, and psychology. The keyword "mhealth" was the dominant term and predominantly used together with the term "physical activity" and "ehealth". A total of three strong clusters of coauthors have been found. Nearly half of the total reported papers were published in three journals. The United States, the United Kingdom, and the Netherlands have the highest degree of author collaboration and a strong institutional network. Mobile phones were most often used as a technology platform, regardless of the targeted behavioral domain. Physical activity and healthy eating were the most frequently targeted behavioral domains. Most articles did not report about the behavior change techniques that were applied. Among the reported behavior change techniques, goal setting and self-management were the most frequently reported. CONCLUSIONS: Closer cooperation and interaction between behavioral sciences and technological areas is needed, so that theoretical knowledge and new technological advancements are better connected in actual applications. Eventually, this could result in a larger societal impact, an increase of the effectiveness of digital technologies for health behavioral change, and more insight in the relationship between behavioral change strategies and persuasive technologies' effectiveness.

A Validation Study of the Fitbit One in Daily Life Using Different Time Intervals.

PURPOSE: Accelerometer-based wearables can provide the user with real-time feedback through the device's interface and the mobile platforms. Few studies have focused on the minute-by-minute validity of wearables, which is essential for high-quality real-time feedback. This study aims to assess the validity of the Fitbit One compared with the ActiGraph GT3x + for assessing physical activity (i.e., steps, time spent in moderate, vigorous, and moderate-vigorous physical activity) in young adults using traditional time intervals (i.e., days) and smaller time intervals (i.e., minutes and hours). METHODS: Healthy young adults (N = 34) wore the ActiGraph GT3x+ and a Fitbit One for 1 wk. Three aggregation levels were used: minute, hour, and day. Mixed models analyses, intraclass correlation coefficients, Bland-Altman analyses, and absolute error percentage for steps per day were conducted to analyze the validity for steps and minutes spent in moderate, vigorous, and moderate-vigorous physical activity. RESULTS: As compared with ActiGraph (mean = 9 steps per minute, 509 steps per hour and 7636 steps per day), the Fitbit One systematically overestimated physical activity for all aggregation levels: on average 0.82 steps per minute, 45 steps per hour, and 677 steps per day. Strong and significant associations were found between ActiGraph and Fitbit results for steps taken (B = 0.72-0.89). Weaker but statistically significant associations were found for minutes spent in moderate, vigorous, and moderate-vigorous physical activity for all time intervals (B = 0.39-0.57). CONCLUSIONS: Although the Fitbit One overestimates the step activity compared with the ActiGraph, it can be considered a valid device to assess step activity, including for real-time minute-by-minute self-monitoring. However, agreement and correlation between ActiGraph and Fitbit One regarding time spent in moderate, vigorous, and moderate-vigorous physical activity were lower.

Effectiveness of a medication-adherence tool: study protocol for a randomized controlled trial.

BACKGROUND: Research shows that more than half of the people taking medication for a chronic condition are non-adherent. Nonadherence hinders disease control with a burden on patient quality of life and healthcare systems. We developed a tool that provides insight into nonadherence risks and barriers for medication-adherence including an intervention strategy to overcome those barriers. This study aims to assess the effectiveness of using this adherence tool in starters with cardiovascular or oral blood glucose-lowering medication to improve medication-adherence. METHODS/DESIGN: In a cluster-randomized controlled trial 25 pharmacies in the Netherlands will be randomized to the intervention or control arm. Patients registered in a general practice participating in a collaborative can be included when they start cardiovascular or oral blood glucose-lowering medication prescribed by their general practitioner. Participants complete an assessment consisting of measuring nonadherence risk and potential barriers to adherence. For patients with an increased nonadherence risk, a graphic barrier profile is created, showing to what extent eight cognitive, emotional, or practical barriers are present. All patients will fill in the medication-adherence assessment twice: between 1 and 2 weeks after the start of the medication and after 8 months. The intervention strategy consists of discussing this barrier profile to overcome barriers. Pharmacists and assistants of the intervention pharmacies are trained in discussing the profile and to offer a tailored intervention to overcome barriers. In the control arm, patients receive care as usual. The primary outcome is medication-adherence of patients with a high risk of nonadherence at 8 months follow-up. Secondary outcomes include the difference in the percentage of patients with an increased nonadherence risk between intervention and control group after 8 months, the predictive values of the baseline questionnaire in the control group in relation to medication-adherence after 8 months, medication-adherence after 1 year follow-up, and barriers and facilitators in the implementation of the tool. DISCUSSION: This manuscript presents the protocol for a cluster-randomized clinical trial on the use of an adherence tool to improve medication-adherence. This study will provide insight into the effectiveness of the tool in starters with cardiovascular or oral blood glucose-lowering medication in improvement of medication-adherence. TRIAL REGISTRATION: The Netherlands National Trial Register, NTR5186 . Registered on 18 May 2015.

Effectiveness of personalised support for self-management in primary care: a cluster randomised controlled trial.

BACKGROUND: Self-management support is an important component of the clinical management of many chronic conditions. The validated Self-Management Screening questionnaire (SeMaS) assesses individual characteristics that influence a patient's ability to self-manage. AIM: To assess the effect of providing personalised self-management support in clinical practice on patients' activation and health-related behaviours. DESIGN AND SETTING: A cluster randomised controlled trial was conducted in 15 primary care group practices in the south of the Netherlands. METHOD: After attending a dedicated self-management support training session, practice nurses in the intervention arm discussed the results of SeMaS with the patient at baseline, and tailored the self-management support. Participants completed a 13-item Patient Activation Measure (PAM-13) and validated lifestyle questionnaires at baseline and after 6 months. Data, including individual care plans, referrals to self-management interventions, self-monitoring, and healthcare use, were extracted from patients' medical records. Multilevel multiple regression was used to assess the effect on outcomes. RESULTS: The PAM-13 score did not differ significantly between the control (n = 348) and intervention (n = 296) arms at 6 months. In the intervention arm, 29.4% of the patients performed self-monitoring, versus 15.2% in the control arm (effect size r = 0.9, P = 0.01). In the per protocol analysis (control n = 348; intervention n = 136), the effect of the intervention was significant on the number of individual care plans (effect size r = 1.3, P = 0.04) and on self-monitoring (effect size r = 1.0, P = 0.01). CONCLUSION: This study showed that discussing SeMaS and offering tailored support did not affect patient activation or lifestyle, but did stimulate patients to self-monitor and use individual care plans.

Determinants of successful telemedicine implementations: a literature study.

Telemedicine implementations often remain in the pilot phase and do not succeed in scaling-up to robust products that are used in daily practice. We conducted a qualitative literature review of 45 conference papers describing telemedicine interventions in order to identify determinants that had influenced their implementation. The identified determinants, which would influence the future implementation of telemedicine interventions, can be classified into five major categories: (1) Technology, (2) Acceptance, (3) Financing, (4) Organization and (5) Policy and Legislation. Each category contains determinants that are relevant to different stakeholders in different domains. We propose a layered implementation model in which the primary focus on individual determinants changes throughout the development life cycle of the telemedicine implementation. For success, a visionary approach is required from the multidisciplinary stakeholders, which goes beyond tackling specific issues in a particular development phase. Thus the right philosophy is: 'start small, think big'.

Mobile patient monitoring: the MobiHealth system.

The forthcoming wide availability of high bandwidth public wireless networks will give rise to new mobile health care services. Towards this direction the MobiHealth project has developed and trialed a highly customisable vital signals' monitoring system based on a Body Area Network (BAN) and an m-health service platform utilizing next generation public wireless networks. The developed system allows the incorporation of diverse medical sensors via wireless connections, and the live transmission of the measured vital signals over public wireless networks to healthcare providers. Nine trials with different health care cases and patient groups in four different European countries have been conducted to test and verify the system, the service and the network infrastructure for its suitability and the restrictions it imposes to mobile health care applications.

MobiHealth: ambulant patient monitoring over next generation public wireless networks.

The wide availability of high bandwidth public wireless networks as well as the miniaturisation of medical sensors and network access hardware allows the development of advanced ambulant patient monitoring systems. The MobiHealth project developed a complete system and service that allows the continuous monitoring of vital signals and their transmission to the health care institutes in real time using GPRS and UMTS networks. The MobiHealth system is based on the concept of a Body Area Network (BAN) allowing high personalization of the monitored signals and thus adaptation to different classes of patients. The system and service has been trialed in four European countries and for different patient cases. First results confirm the usefulness of the system and the advantages it offers to patients and medical personnel.

Wireless body area networks for healthcare: the MobiHealth project.

The forthcoming wide availability of high bandwidth public wireless networks will give rise to new mobile health care services. Towards this direction the MobiHealth project has developed and trialed a highly customisable vital signals' monitoring system based on a Body Area Network (BAN) and an m-health service platform utilizing next generation public wireless networks. The developed system allows the incorporation of diverse medical sensors via wireless connections, and the live transmission of the measured vital signals over public wireless networks to healthcare providers. Nine trials with different health care cases and patient groups in four different European countries have been conducted to test and verify the system, the service and the network infrastructure for its suitability and the restrictions it imposes to mobile health care applications.