MBZUAI graduate uses AI for early detection of Alzheimer’s, dementia

Abu Dhabi: With an estimated 75 per cent of dementia cases going undiagnosed globally, Salma Hassan, the valedictorian of the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) Class of 2025, has turned to AI to help change that.

Hassan, who graduated with an MSc in Machine Learning, focused her thesis on using multimodal graph learning for the early diagnosis and prognosis of dementia and Alzheimer’s disease – a field she describes as under-explored but carrying enormous human impact.

“It is estimated that every three seconds, someone is diagnosed with dementia – though 75 per cent of cases are never even diagnosed, let alone treated. It is a significant global issue, and I saw an opportunity for AI to bridge this gap and improve the situation,” the 23-year-old said.

Her pioneering research seeks to improve diagnostic accuracy and identify early biomarkers by leveraging relationships across data types, including brain imaging, genetics, and clinical records – modelling how different biological and structural features interact.

Her work highlights differential diagnosis, identifying not only whether a patient has dementia, but also which subtype they may have – a distinction that is critical for treatment but often difficult to make.

“It is very hard to differentiate between them, but it is important, and machine learning can help with this.”

The second aspect of her research delved into prognosis, focusing on predicting cognitive decline and estimating when patients in the preclinical stage, those without symptoms, might progress to full-blown dementia.

Hassan noted that her goal was to build tools that provide clarity to clinicians, enabling smarter allocation of medical resources, especially in low- and middle-income countries, where access to specialists is limited.

Hailing from Egypt, Hassan joined MBZUAI in 2023 after graduating early from the American University of Sharjah with a degree in computer engineering. Her undergraduate project involved building a brain-computer interface virtual reality system to support rehabilitation in patients with upper limb impairments – a project that first sparked her interest in AI's medical applications.

Guided by her advisor, Dr Mohammed Yaqub, Hassan has already published her findings in Nature Scientific Reports and presented at major conferences such as ICIP and MICCAI.

Looking ahead, Hassan aims to expand her research to include genetic markers – an area she believes holds the key to even earlier detection. Accepted into the MBZUAI PhD programme, she remains committed to developing AI-powered diagnostic tools that are scalable, interpretable, and clinically useful.

“I feel like I’m just getting started, and there’s still a lot more to do,” she said.