This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR mhealth and uhealth, is properly cited. The complete bibliographic information, a link to the original publication on http://mhealth.jmir.org/, as well as this copyright and license information must be included.
Medical smartphone apps and mobile health devices are rapidly entering mainstream use because of the rising number of smartphone users. Consequently, a large amount of consumer-generated data is being collected. Technological advances in innovative sensory systems have enabled data connectivity and aggregation to become cornerstones in developing workable solutions for remote monitoring systems in clinical practice. However, few systems are currently available to handle such data, especially for clinical use.
The aim of this study was to develop and implement the digital health research platform for mobile health (DHARMA) that combines data saved in different formats from a variety of sources into a single integrated digital platform suitable for mobile remote monitoring studies.
DHARMA comprises a smartphone app, a Web-based platform, and custom middleware and has been developed to collect, store, process, and visualize data from different vendor-specific sensors. The middleware is a component-based system with independent building blocks for user authentication, study and patient administration, data handling, questionnaire management, patient files, and reporting.
A prototype version of the research platform has been tested and deployed in multiple clinical studies. In this study, we used the platform for the follow-up of pregnant women at risk of developing pre-eclampsia. The patients’ blood pressure, weight, and activity were semi-automatically captured at home using different devices. DHARMA automatically collected and stored data from each source and enabled data processing for the end users in terms of study-specific parameters, thresholds, and visualization.
The increasing use of mobile health apps and connected medical devices is leading to a large amount of data for collection. There has been limited investment in handling and aggregating data from different sources for use in academic and clinical research focusing on remote monitoring studies. In this study, we created a modular mobile health research platform to collect and integrate data from a variety of third-party devices in several patient populations. The functionality of the platform was demonstrated in a real-life setting among women with high-risk pregnancies.
The increasing penetration of smartphones into consumer markets, as well as the growth in connected devices for health care, sport, and wellness, is leading to a dramatic increase in consumer-generated data [
It is already possible to connect mobile health apps to a wide range of portable devices or sensors [
Remote monitoring patients at home requires a data communication system between the patient and the health care professional. Typically, the patient is equipped at home with a set of sensors that measure vital signs and a smartphone app that collects data on behavior, symptoms, and location. These data are collected on a cloud-based server that is accessed by the health care professional. Tools for decision support to assist the health care professional and semiautomatic feedback to patients are necessary to manage large patient populations.
Despite the significant technological advances in developing novel sensory systems and state-of-the-art devices, there has been limited investment in developing the infrastructure that is required to connect and handle the amount of information that these devices generate. In particular, there are limited tools available to handle this health information in terms of clinical applications [
In this paper, we discuss the development of a generic and open digital research platform for remote monitoring that can be used to perform academic and clinical research into remote monitoring. This digital platform overcomes the problems of third-party device integration and collection of various data feeds from patient populations. The collected data can be analyzed using the platform and presented visually to the caregiver. We demonstrate the functionality of the platform in a real-life setting, among women with high-risk pregnancies.
The concept and design of our modular digital health research platform for mobile health (DHARMA) platform were focused primarily on clinical usability. Accordingly, health care professionals (doctors and nurses) participated in each stage of its conceptualization and development, from design and prototyping through to implementation and user trials.
The DHARMA platform was developed in close collaboration with all stakeholders: doctors, professional caregivers, researchers, and experts in mobile health. As its primary aim is to support clinical studies, the platform was built taking into account the requirements of specific studies but always with general applicability and genericity in mind. We ensured short test cycles by using the agile development methodology (Scrum). An appropriate toolset was put into place for managing user stories (Jira), documenting design and implementation decisions (Confluence), and continuous build and deployment (Gitlab). To modularize implementation, the GitFlow approach was used. To separate continuous building and release versions, deployment was split up into a development, test, and production environment. Teams collaborated efficiently using Slack. For components of specific studies, the core development team was extended by internship students.
The app comprises a Web-based user interface and an object-oriented programming language (Java)–hypertext preprocessor (PHP) back end. The platform is set up on a cloud server. The front and back ends are coded in various languages. The front end is built with PHP with a combination of bootstrap and Laravel frameworks. The back end is built with a combination of PHP and Java to ensure cross-platform compatibility. The server runs on Windows, Apache, MySQL, and PHP platforms. Owing to the highly valuable content of the database system, the database is backed up daily and archived.
The platform was built as a pluggable component-based middleware. As mentioned by Piwek et al [
The concept of components was introduced to manage different studies on a single platform. Each study comprises at least 1 or multiple components (eg, data handling and questionnaires) that can be activated by a study leader.
Every clinical study is divided into multiple levels, starting with the hospital acting as the lead partner of the clinical study (level 1; eg, Hospital East-Limburg, Genk, Belgium). The next level is the medical domain (level 2; eg, gynecology), and the third level is the study itself (level 3; eg, pregnancy remote monitoring study, Premom). This multilevel approach allows us to implement user rights and develop alert thresholds on each level of the study. Currently, 3 user profiles are defined: patient, study leader, and administrator (admin). After completing the registration process, 2 possible methods are provided. First, the admin can appoint a user as a study leader, and then a study leader can assign the user as a patient for inclusion in a specific study. This hierarchical model allows study leaders to independently create and follow-up patients.
The goal of DHARMA is to aggregate and visualize multiple vital parameters collected by using devices and medical apps from different vendors without the need to consult the vendor-specific platforms. The remote monitoring platform can receive information directly or collect data by connecting to other databases. If new values are uploaded by a user, DHARMA receives a notification and automatically launches the technical process needed to aggregate and store the data in its own database. Duplication of these data enables secure storage and accessibility for analysis and alert generation.
To generate an efficient data collection workflow, an alert engine was developed, which can interpret and handle medical and technical alerts. Medical alerts are based on the collected data and detect a value outside the specified thresholds. The thresholds are specified for each study (level 3) and can dynamically support longitudinal changes and more complex interpretations. However, the configuration of patient-level alerts can be set to individual ranges based on clinical guidelines. On the basis of the clinical input, the alerts are categorized as normal, medium, or high priority. Technical alerts are defined by messages containing information about missed data transmissions. This triage system could help organize the clinical call center activities.
Overview of the data flow of the digital health research platform for mobile health platform.
The
A patient file component was created to arrange the information into individual patient records. Each record comprises 4 main tabs: medical information, statistics, questionnaires, and follow-up. The medical information tab allows the study leader and the clinician to view study-specific patient parameters or comorbidities. Study parameters include the study-specific information needed to interpret the vital parameters. The statistics tab displays a graphical overview of each vital parameter. The questions tab provides an overview of the questionnaires that were sent to and completed by the patient. The follow-up tab allows caregivers to send text messages among multiple disciplines. Each patient contact is logged in this tab.
A report component was created to provide a comprehensive digital overview of the patient’s status. The overview can be printed or downloaded and emailed to the patient’s doctor or caregiver. Lava Charts (Google Chart API) was used to visualize the patient’s data in charts and graphs.
The remote monitoring study platform was built between February 2015 and July 2018. The custom-made remote monitoring platform was deployed to monitor patients outside the hospital in several studies, including multiple sclerosis, low back pain, and osteoporosis studies in which the patients’ activity data (number of steps and intensity) were tracked.
Remote studies included in digital health research platform for mobile health.
Study | Sample size, n | Blood pressure, n | Weight, n | Activity data, n |
Premom | 604 | 95,835 | 9430 | 35,520 |
Multiple sclerosis | 36 | 0 | 14 | 1544 |
Low back pain | 33 | 12 | 110 | 2191 |
Osteoporosis | 13 | 25 | 857 | 1603 |
The main study in which DHARMA was tested was the Premom [
Besides collecting and handling different sources of information, one of the main objectives of the platform is to provide the researcher/clinician with efficient visualization of all patient data. If the data triggered a specific alert, the dashboard prioritized the alerts based on the predefined thresholds and displayed them to the person responsible for reviewing the data. This enables the platform to triage patient alerts and facilitate patient handling and follow-up.
Screenshot of the alert representation upon login of the dashboard. Data was triaged based on predetermined thresholds into High Risk, Medium Risk or Normal.
The patient records bundle the individual patient’s information into a single file. The received parameters are individually plotted as functions of time to identify specific trends that could trigger an alert by crossing specified thresholds. For the patient shown in
Overview graph of a patient’s blood pressure displaying both the systolic and diastolic blood pressure, with the predefined alert thresholds indicated with dashed lines.
The challenges of working with different types of data include how to handle, analyze, and store the data appropriately and transparently. In the normal workflow, only the summarized values of larger datasets are used because granular detail is not required for daily patient management. However, if required, the data are available for Web-based data processing or can be exported for offline scientific research, such as the development of novel algorithms or data-processing techniques.
An example of longitudinal activity data for a period of about 2 months.
Granular example of the activity level per minute for a period of about 15 hours.
This paper outlines the development of a digital health research platform for remote monitoring. By combining advanced wearable sensors with smartphone technologies for remote monitoring, it is possible to monitor the health of patients in their home environment, an approach that may reduce the number of
A health monitoring system can only provide its greatest usability if it can be fully integrated into the user’s and the physician’s daily workflow. The goal of our study is to integrate the data streams from multiple medical device vendors and allow health care practitioners to efficiently evaluate a patient’s health status. This will improve efficiencies in cost and time. In addition, the platform was designed to enable rapid and cost-effective scalability.
Privacy is a fundamental right in the public health care domain, especially following the recently implemented GDPRs. Health care practitioners and patients are becoming increasingly aware of this important aspect. Confidential handling and storage of private patient data have also become a critical aspect of study design. Therefore, all personal data in our platform are deidentified and every unique identification number, characteristic, or name is removed. Moreover, all participants need to provide signed written informed consent.
The development of a centralized visualization platform has been described in earlier reports, for example, for monitoring arrhythmias [
Although DHARMA provides exciting opportunities to improve remote monitoring services, it is not free from limitations. First, the data recorded by the medical devices are initially sent to the vendor’s dedicated database. This means that the vendor (eg, iHealth or Withings) also owns the patient’s data. An iHealth/Withings study profile without the patient’s personal data was created in the Premom study to deidentify the patients who were included in the follow-up program. This approach could be improved by creating a third-party app that connects directly to the medical devices; however, not all vendors allow direct access to their medical device via an open API. This process would bypass data transfer to an external company. A second limitation is the applicability of remote monitoring studies among technophobic individuals and people with limited cognition or ability to express consent, such as neonates, elderly, and sedated patients in an intensive care unit [
Currently, participants in DHARMA remote monitoring studies need to provide written informed consent to be enrolled in each study, as previously reported by Eysenbach et al [
Another step to improve the platform will involve embracing the definitions of standard information models and information technology communication standards, such as Health Level 7 fast healthcare interoperability resources, together with clinical terminologies, such as systematized nomenclature of medicine—clinical terms, to ensure interoperability with hospital electronic medical record (EMR) systems [
Smartphone health apps and medical devices collect large amounts of vendor-specific data. There are currently very few tools to collate and handle the data generated by multiple medical devices. We developed a component-based digital research platform to integrate the data in different formats from different sensors into a single integrated system. The platform performed well in a health care setting in real-time circumstances for the follow-up of pregnant women at risk of developing pre-eclampsia. The next stage in its development will involve integrating the platform with existing EMR systems to create a closed-loop information system.
Scientists or companies willing to contribute to this study are welcome to contact the authors.
advanced encryption standard
app programming interface
digital health research platform for mobile health
electrocardiogram
electronic medical record
general data protection regulation
personal computer
This study is part of the Limburg Clinical Research Program UHasselt-ZOL-Jessa that is supported by the Limburg Sterk Merk Foundation, the province of Limburg, the Flemish government, Hasselt University, Ziekenhuis Oost-Limburg, and Jessa Hospital.
None declared.