Keywords
Clinical Conditions
Equipment & Techniques
My research expertise spans both applied and theoretical AI, including Deep Learning, Multi-Modal AI, and eXplainable AI. Throughout my career, I have collaborated with interdisciplinary teams to tackle challenges in biomedical applications, and healthcare. At the University of Cambridge, I am working on foundational AI models for medical imaging, leveraging deep learning and explainability for diagnostic and prognostic predictions. Over the past several years, my research has focused on leveraging state-of-the-art AI techniques to solve complex biomedical problems and neuroscience. My work has encompassed developing XAI models for clinical imaging and genomics that directly inform diagnostic and therapeutic decisions. For example, in collaboration with multi-disciplinary teams from Oxford, Glasgow, Lincoln, Brighton, and Sheffield, I have contributed to biomarker discovery in COVID-19 by integrating phenotyping, blood, and imaging datasets. My collaborative projects at the University of Cambridge have further honed my expertise. Working with Professor Richard Clayton’s group (University of Sheffield, till 2022), I developed robust computer vision AI applications for cardiac arrhythmias, while my engagements with Professor Andriew Shift’s group i(University of Sheffield, till 2022) involved pulmonary hypertension research focusing on uncertainty estimation and explainability. Additionally, I have collaborated with Professor George Panoutsos (University of Sheffield, till 2022) on metastatic mutation cell detection and the explanation of potential biomarkers, as well as with Professor Pietro Lio, Murray Graham and John Suckling (University of Cambridge, till 2025) in neuroscience, focusing on secondary sulci region identification, phenotyping, and genetic correlations in psychotic patients and brain tumors. Currently, I am actively involved in several cutting edge projects that align closely with the mission of the University of Cambridge and its School of Clinical Medicine. I am developing predictive models for neuroscience by integrating multimodal data including text, imaging, phenotyping, and genomics, using explainable AI on large scale datasets. Notably, I am also engaged in research involving pre and post surgery datasets from brain tumour patients, applying gradient based explainability methods to uncover novel connectomic patterns in high grade glioblastoma. Additionally, I serve as a Senior Research Associate at the University of Cambridge School of Clinical Medicine and the Cancer Research UK (CRUK) Cambridge Centre, where I focus on childhood brain cancer.
3D eXplainability Artificial Intelligence
The 3D explainable framework that provides both local and global interpretations and explanations of our deep learning 3D classification network's results. The ratio of the aithfulness and complexity metrics are compute in that stage. In this example we include only the GradCam explainability method just for simplicity.
