Abu Dhabi: Here's how MBZUAI’s Indian PhD student is solving a major problem in medical AI

Abu Dhabi: Umaima Rahman, a 29-year-old researcher from India, is striving to make artificial intelligence (AI) in healthcare safer, fairer, and more reliable, no matter where it’s used.
This week, she received her PhD in Computer Vision from the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), becoming one of the first female students to earn a doctoral degree from the university.

AI system that works across hospitals

Rahman’s research focuses on a major problem in medical AI, i.e., AI systems often fail when moved from one hospital to another. That’s because the machines and imaging methods used in different places can vary a lot. For example, an AI tool trained on X-rays from a modern hospital may not work well in a clinic with older equipment and this could lead to harmful misdiagnoses.
“Imagine a world where an AI model, trained to detect cancer from medical images with near-perfect accuracy, suddenly fails when deployed in a different hospital,” she said.
“The issue? A slight change in imaging protocol, and the model built on data from another institution can no longer perform.”

Safer, fairer AI for real-world settings

To solve this, Rahman spent four years developing AI models that can ‘generalise’ or work well across different hospitals, scanners, and patient types. Her models learn to focus on important medical features, like signs of a disease, while ignoring differences in image quality or scanner type.
She stressed that in healthcare, poor AI performance can harm people, and hence the need for systems that are safe and reliable everywhere.

“In medical imaging, where the stakes are literally life and death, this isn’t just an academic problem, it’s a matter of patient safety and equitable healthcare delivery.”

New method aids rare and emerging illnesses

Rahman also introduced a concept called “cross-disease transferability,” where AI trained to detect one disease can help identify others in the same organ – a helpful tool during health crises like COVID-19. Her work is especially useful for countries with limited resources, where high-end machines aren’t always available.

Rahman presented her work internationally, including in Switzerland, where she met medical experts and AI researchers.

‘UAE feels like home now’

Now exploring postdoctoral opportunities at global institutions like Stanford and MIT, Rahman hopes to eventually return to the UAE and continue her mission.
“AI isn’t here to replace doctors. Instead, our goal is to develop AI methodologies that assist and empower clinicians, supporting their decision-making processes and enhancing the quality of care.”
Rahman also hopes to inspire more women to join the field.
“I’d love to become a professor one day. There are a lot of women in biomedical engineering, but not in AI. I hope that changes.”