Scientists from the Qatar Computing Research Institute (QCRI) at the Hamad Bin Khalifa University (HBKU) have conducted an investigative trial on the impact of fasting during the holy month of Ramadan on diabetic patients.
Working in collaboration with physicians from Hamad Medical Corporation (HMC), the trial informed ongoing research into the use of Artificial Intelligence (AI) in caring for diabetic patients wishing to fast.
More than 60 patients with Type 2 diabetes were provided with a Fitbit to measure activity and sleep, as well as a wearable device that continuously monitored glucose readings prior to and during the holy month.
QCRI’s System for Integrated Health Analytics was also piloted to ingest, visualise and analyse data, with additional AI models developed to predict possible hyper- and hypoglycemic events.
The trial results informed a paper accepted at the ‘Journal of Diabetes Research and Clinical Practice’, which concluded that XGBoost, a machine learning AI algorithm, achieved high predictive performance for normal and hyperglycemic excursions.
The same application nevertheless offered limited predictive value for hypoglycemia in patients on multiple therapies who fasted during Ramadan, HBKU said in a statement.
Dr Faisal Farooq, principal scientist and head of the Center for Digital Health and Precision Medicine, QCRI, said: “It was an honour to collaborate with HMC on a project that undoubtedly resonates with our commitment to tackling computing challenges that positively touch upon people’s lives.
“The AI models developed to inform and support the trial have shown their potential to save lives by predicting low blood sugar events during fasting. They also provide physicians with key insights into risk factors that cause these events, such as a patient’s treatment regimen. Doing so raises the prospect of protocols that take into account local cultures and traditions, particularly when it comes to the observance of religious practices.”
Dr Faisal Farooq
