Leveraging state-of-the-art techniques in artificial intelligence (AI), scientists at the College of Science and Engineering (CSE) at Hamad Bin Khalifa University (HBKU), have proposed a novel AI-enabled system that can be used to diagnose cardiovascular diseases (CVDs) based on retinal (innermost, thin, light sensitive layer of eye) images and bone health data.
This seminal method was developed in-house at CSE, HBKU and was funded by Qatar Biobank (QBB). The research findings were published in the Open Access Sensors journal which is publicly available for the community.
CSE's proposed deep learning-based system diagnoses CVDs using combined information from two imaging techniques. The first, retinal images, provides medical practitioners with a fast way to examine optic nerves and diagnose diseases such as hypertension, diabetic retinopathy, or arteriosclerosis. The second is dual-energy X-ray absorptiometry (DXA) data on bone mineral density. DXA is a form of non-invasive X-ray technology used to analyze bone health and bone loss. DXA has been approved recently for use in CVD diagnosis by the United States' Food and Drug Administration (FDA).
The investigators leveraged deep learning techniques to fuse the two sets of information (retinal imaging plus DXA data) to diagnose the onset of CVD in a sample group of Qatari adults. Based on the experiment results, the proposed novel method demonstrates major potential advancement for the early detection of CVD in a fast, non-invasive, and low-cost manner.
In Qatar, 69 percent of mortalities are caused by chronic diseases where 24 percent are caused by CVDs. CSE's research is in line with the Qatar National Health Strategy 2018-2022 which aims at reducing the mortality rate caused by diseases such as CVD by 15 percent. One way of achieving this is by utilizing precision or personalized health for its early detection and diagnosis.
Dr. Tanvir Alam, Assistant Professor in the Information and Computing Technology Division at CSE, and the project's lead principal investigator, said: "At the heart of this study is the use of state-of-the-art AI techniques to prevent, diagnose, and treat cardiovascular disease. The results demonstrate how the use of AI in healthcare, especially deep learning-based models, is evolving and its exciting potential to be expanded in clinical settings. The study has been enriched by the joint efforts of our research collaborators, experiments by HBKU PhD student Hamada Al-Absi, and cooperation with Qatar Biobank."
Dr. Mounir Hamdi, Founding Dean of CSE, HBKU commented on the study outcomes: "This kind of seminal, innovative research conducted at CSE, will enable precision medicine solutions for Qatari society and elevate Qatar's leadership in the field of AI and personalized healthcare."
(L-R) Dr. Tanvir Alam, Dr. Mounir Hamdi, Hamada R. H. Al-Absi