Open Wearables Initiative Promotes Use of Sensor-Generated Measures in Clinical Research
Several stakeholders in wearable health technology have formed the Open Wearables Initiative (OWEAR) to promote the effective use of high-quality, sensor-generated measures of health in clinical research through the open sharing and benchmarking of algorithms and datasets.
Wearables, ingestible sensors and in-home monitoring technologies offer the opportunity to assess an individual's health continuously, objectively and in real time. However, the initiative notes that the lack of accepted endpoints is proving to be a major impediment to the adoption of these digital measures in clinical trials. (For example, reducing blood pressure is the primary endpoint for hypertension control.)
OWEAR will leverage the work of thousands of researchers from academia, pharma, and other organizations during the past decade to facilitate the development of those requisite endpoints.
OWEAR is now actively soliciting open source software and datasets from wearable sensors and other connected health technologies. OWEAR has also expanded its Working Group to include executives from four major pharmaceutical companies, clinical research organization Sage Bionetworks and the Digital Medicine Society (DiMe).
OWEAR will serve as a community hub, indexing, distributing and benchmarking algorithms openly and transparently. It will act as a neutral broker, conducting formal, objective benchmarking processes and identifying high-performing algorithms in selected domains. Its goal is to provide the industry with a searchable database of benchmarked algorithms and source code that can be freely used by everyone. The goal is to help streamline drug development and enable digital medicine.
OWEAR was co-founded in Sept. 2019 by Shimmer Research; Dr. Vincent van Hees, author of the GGIR software and algorithms for movement sensor calibration, sensor wear detection, and signal aggregation; and Nextbridge Health, which is developing the Nextbridge Exchange, an online marketplace and discovery platform for the clinical research community.
In a conscious effort not to duplicate resources, OWEAR will not host the software or datasets because that role is already fulfilled by repositories such as GitHub, Synapse.org, and UCI Machine Learning Repository. OWEAR is focused instead on aggregating all the available resources and collecting the required metadata into an index so that healthcare researchers can more easily find and evaluate the algorithms and software that best meet their needs.
OWEAR is asking software developers and medical researchers to register algorithms related to digital medicine and publicly available datasets from wearables and connected health technologies, even those with license restrictions, at http://www.owear.org. This information will be collected into an index of all current resources, which will be made available soon.
By registering their algorithms and datasets with OWEAR, developers will gain recognition for their commitment to advancing the field. Contributors will also gain greater visibility for their work among clinical researchers in academia and industry, leading to potential collaboration and consulting opportunities. They will also receive feedback and potentially help from end-users and other developers to augment their work.
