Leaders in AI-Driven Health Research

There are three main modes for research done within my group:
1) Development of new computational methods and tools for dealing with single cell omics data. For this we typically use publicly available data
2) Collaboration with other groups and analysis of data generated by them, or to which we can get access, such as the Emirati Genome Program or the Human Phenotype Project
3) Collaborations with clinicias or other institutional partners to design and execute novel clinical and biological studies, generate data, and analyze this data for biological and clinical discovery.

• Early immune reconstitution assessment by flow/mass cytometry in autologous hematopoietic transplantation performed in a patient with multiple myeloma.
• Recurrent clumping in the extracorporeal photopheresis circuit using acid citrate dextrose solution A.
• Collection efficiency of double- versus single-needle apheresis in mononuclear cell therapy manufacturing.Mononuclear cell recruitment during extracorporeal photopheresis: Partial results of a phase 1/2 randomized clinical trial in multiple sclerosis.
• A randomized clinical trial, Abu Dhabi, 2020 – insights into mononuclear cell mobilization in autoimmune diseases.
• Partial results of a phase 1/2 randomized clinical trial in multiple sclerosis – mononuclear cell recruitment during ECP.
• First review of multiple myeloma patients in Sheikh Khalifa Medical City, Abu Dhabi, United Arab Emirates.

We mine large data sets of genetic variation linked to health and disease outcomes to build inference of gene::disease associations as well as map genetic variants in the human genome underlying these associations (alone or in combination). We are interested in the problem on a global scale and seek to understand why certain populations may exhibit increased (or decreased) risk of disease based on a combination of genetic, environmental, and cultural factors. We have led development of novel AI/ML tools for this kind of work and collaborate broadly with other groups and centers doing this type of work.

We combine data-driven computational analysis, genomics, machine learning, and cloud-scale bioinformatics infrastructure to interrogate cancer genomes both in depth and at scale. Our research is highly collaborative – bringing together clinicians, biologists, data scientists, and global consortium partners to generate rigorous, reproducible, and clinically meaningful insights. Grounded in the principles of open science and reproducibility, we are committed to building inclusive research capacity by training the next generation of scientists, with a special focus on empowering communities in the UAE and the Global South.

We develop AI models to model the biological system properties, predict biomoleculare status and system response, discover new functional molecules, and develop new drug and treatment to cure diseases. We collaborate with both biologists and clinicians, academia and industry.

My research strategy employs integrative computational methods to bridge molecular science and public health. I specialize in developing strategies for multi-omics data integration, combining genomic, transcriptomic, and epigenomic information with clinical data. My approach prioritizes methodological frameworks that ensure equitable performance across diverse populations. A key focus involves using analytical and data-driven techniques to identify fundamental disease drivers beyond simple correlations. The ultimate goal is to translate these computational discoveries into clinically actionable tools, creating a pathway from multi-omics insights to meaningful public health impact.

The strategic focus of Prof. Pržulj’s multi-disciplinary team is two-fold: 1) to exploit the existing, deep-tech technologies developed in her lab and transfer them from academia to industry for the benefit of the society; and 2) to pursue the development of new, forward-looking, cutting-edge computational and bio-tech technologies that will provide qualitatively new ideas toward changing of the existing paradigms and enabling a quantum leap in biomedical advancement. The key for both is to recruit, train and retain global, multi-disciplinary talent, motivated to witness real-world impact of their research.

My research focuses on practical needs and challenges in medicine and healthcare. I collaborate with clinicians, scientists, and engineers to conduct interdisciplinary research that creates insights and solutions for advancing global health innovation and medical breakthroughs.

I take a collaborative, data-driven approach that integrates advances in machine learning with deep phenotyping from the Human Phenotype Project. My work emphasizes designing modular, interpretable methods and validating them across multiomics, imaging, wearables, and lifestyle data. I partner with clinicians and domain scientists to ensure discoveries translate into meaningful biological and health insights.

Imran’s research combines AI-driven modeling with interdisciplinary collaboration across medicine, biology, machine learning and computer vision to enable early disease diagnosis and improve patient outcomes. He works closely with clinicians and researchers to translate advanced AI methods into real-world healthcare solutions, ensuring that models are interpretable, ethical, and clinically relevant. His work focuses on integrating multi-omics and clinical data to generate actionable insights that advance precision medicine.

Segal’s research strategy emphasizes large-scale, deeply phenotyped human cohorts coupled with advanced AI and statistical modeling. He founded the Human Phenotype Project, a longitudinal multi-omic cohort integrating genomics, transcriptomics, continuous monitoring, imaging, and clinical data to enable predictive modeling of human health. His approach focuses on developing foundation AI models trained on multimodal human data to predict disease onset, metabolic dynamics, aging, and individual health trajectories. Methodological rigor, scalable data integration, and translational relevance guide both his computational and experimental research programs.

My research leverage deep learning–based AI technologies to model complex biological processes, such as gene regulation and antigen presentation.

Professor Ahsan leads multidisciplinary teams to build and leverage diverse population and patient cohorts and integrate multimodal data to develop scalable, equitable, and real-world AI-driven tools for precision medicine. By bridging population-scale biomedical and healthcare data science and AI tools with genomic and clinical epidemiology, he aims to shape the future of precision health solutions to improve disease prediction and prevention.