Ikboljon Sobirov

Supervised by Charalambos Antoniades 

A researcher investigating how biological information is encoded in medical images and how AI can recover, quantify, and interpret those signals.

Ikboljon Sobirov is an AI scientist and DPhil candidate at the University of Oxford. His research examines how biological information is encoded in medical images and how computational methods can recover, quantify, and interpret these signals to advance precision medicine. By integrating cardiovascular imaging, multi-omics data, electronic health records, and machine learning, he develops imaging-derived biomarkers that offer a non-invasive window into disease biology.

His current work focuses on cardiovascular disease, where he studies how imaging can be used not only to predict clinical outcomes but also to reveal the biological mechanisms underpinning disease progression. Through approaches spanning radiomics, radiotranscriptomics, and multimodal learning, he seeks to transform medical imaging from a largely descriptive tool into a quantitative measure of human biology.

Alongside his doctoral studies, Ikboljon is a Research Fellow at Caristo Diagnostics, where he helps develop clinically deployable AI technologies. His work focuses on building robust, interpretable systems that connect biological discovery, computational innovation, and real-world clinical application.

Before joining Oxford, Ikboljon completed an MSc in Computer Vision at the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), with his research focused on medical imaging and cancer prognosis. He also holds a BSc (Hons) in Business Information Systems from Westminster International University in Tashkent.

His long-term goal is to establish imaging as a quantitative measure of human biology, enabling the discovery of biomarkers that link anatomy, molecular processes, and clinical outcomes, thereby supporting earlier intervention and more personalised healthcare.