A group of Qatar-based researchers have developed an eHealth platform –System for Integrated Lifestyle Health Analytics (Siha) - to collect data from wearables for clinical use. This has created a new pathways for clinicians to access data for various medical purposes.
With copyright in place, it is currently available for licensing.
“Lifestyle data captured by wearables can be particularly useful in chronic disease management, but the challenge has been that clinicians don’t have a way to access it,” said Ummar Abbas, senior software engineer at Qatar Computing Research Institute (QCRI), part of Qatar Foundation’s (QF) Hamad Bin Khalifa University.
According to an article on the QF website, most hospitals and clinics use Electronic Health Record (EHR) systems to store patient history. However, EHR-based systems are yet to integrate data from wearables.
“It doesn’t happen anywhere in the world and the reason for that being most consumer-based wearable devices are not US Food and Drug Administration approved. This is slowly beginning to change though,” he pointed out.
“The first thing we need is consent from a user to collect their data, and once we have that, the user is instructed to download an app on their phone and this app then fetches the data from their device and sends it to our cloud, which is based in Qatar, where it is stored in a secure encrypted database,” Abbas said. He revealed the process isn’t as simple as it sounds because of differences in the way different wearables collect and store data.
“Some wearables upload the data to their cloud; others store it on the device itself,” the researcher said.
Another challenge is each device has its own data format and collection frequency, so the team’s first challenge was harmonising data formats from different brands and presenting a unified interface to the clinicians.
In addition to the data being used by clinicians to devise treatment plans and track lifestyle changes, it is also being used to build and train AI models.
Syed Hashim, a software engineer at QCRI and a member of the Siha development team, noted: “What these models do is predictive analysis. For example, the team is currently working on a model which will use wearable data and data from diabetic patient’s continuous glucose monitor to predict the occurrence of low blood sugar before it happens and can send intelligent notifications to the user to take corrective actions.”
The scientist said that having access to data from wearables is particularly useful in managing conditions that require a lifestyle change, for example, diabetes or cardiovascular diseases.
“Having access to wearable data will allow clinicians to quantitatively track changes in the lifestyle of their patients. It also adds a sense of accountability for the patient because they know the doctor has access to their data and can tell whether the patient is adhering to recommendations,” he highlighted.
Currently, Siha supports only wearables by Apple, Google, Fitbit, Huawei, and Withings. “These brands were chosen because of the extensive research they put in to ensure the accuracy of their features, and their constant efforts to further improve their accuracy,” pointed out, Abbas.
Siha is currently being tested in various clinical trials locally, including one for diabetes in adults at Hamad Medical Corporation and one for asthma in children at Sidra Medicine.
According to Abbas, another advantage of wearables is that they record data even when the user is healthy, “essentially building a personalised physiological health model and baseline for each individual”, which in turn makes it easier to spot anomalies and know when something is unusual, such as the onset of an illness or worsening of a chronic condition.
“We would like to see our system being integrated into regular clinical EHR systems and be a part of the regular hospital workflow in the country. When a patient’s records were to be pulled up, in addition to the vitals taken at every doctor's visit, you would also see the data from their wearable device,” Abbas, added.