I’m currently a research scholar at the Mass General Cancer Center working on building agentic solutions for radiotherapy treatment planning with Dr. Raymond Mak. Radiotherapy treatment planning lies at the convergence of physics, biology and computation making it an interesting and impactful use-case to bring agents to.

I’m in the final year of my PhD (waiting to defend my thesis) at the AIM program housed at Mass General Brigham and Harvard Medical School where I largely focused on developing representation learning methods for cancer imaging and oncology, advised by Prof. Hugo Aerts. Amidst the pandemic, I graduated cum laude with a MSc in Artificial Intelligence from the Department of Advanced Computing Sciences at Maastricht University and worked on AI-assisted radiotherapy at MAASTRO Clinic. I come from an engineering background (Bachelors in Electronics and Comms Engg.) with skills honed through freelancing and roles at several startups.

With my experiences, I hope to build AI and engineering systems that can benefit medical care.

Whenever I get a chance, I enjoy spending time in nature (New England is a birders paradise), meditation and digging deeper into topics in metaphysics, ethics, epistemology and world history.

🔥 News

• 2025.01:   Our preprint on developing vision-centric foundation models for CT is out!
• 2024.05:  🎖️ Our challenge won ‘Best in Physics’ award at ESTRO2024
• 2024.03:  🎉📝 Our paper on foundation models for cancer imaging biomarkers was published in Nature Machine Intelligence
• 2022.02:  🎉📊 Our paper on design choices for CycleGANs in CBCT to CT translation will be presented at SPIE in San Diego by Ibrahim

📝 Publications

Representation Learning in Radiology

• arxiv [Foundation Model for Cancer Imaging Biomarkers](https://www.nature.com/articles/s42256-024-00807-9) Suraj Pai, Ibrahim Hadzic, Dennis Bontempi, Keno Bressem, Benjamin H. Kann, Andriy Fedorov, Raymond H. Mak, Hugo J. W. L. Aerts

Representation Learning in Oncology

• Nature Machine Intelligence Foundation Model for Cancer Imaging Biomarkers Suraj Pai, Dennis Bontempi, Ibrahim Hadzic, Vasco Prudente, Mateo Sokač, Tafadzwa L. Chaunzwa, Simon Bernatz, Ahmed Hosny, Raymond H. Mak, Nicolai J. Birkbak & Hugo J. W. L. Aerts

Image-to-Image translation for Medical Imaging

• SPIE 2023Optimizing CycleGAN design for CBCT-to-CT translation: insights into 2D vs 3D modeling, patch size, and the need for tailored evaluation metrics Suraj Pai, Ibrahim Hadzic, , Vicki Trier Taasti, Dennis Bontempi, Ivan Zhovannik, Richard Canters, Jan Jakob Sonke, Andre Dekker, Jonas Teuwen, Alberto Traverso
• Sensors Frequency-Domain-Based Structure Losses for CycleGAN-Based Cone-Beam Computed Tomography Translation Suraj Pai, Ibrahim Hadzic, Chinmay Rao, Ivan Zhovannik, Andre Dekker, Alberto Traverso, Stylianos Asteriadis, Enrique Hortal

3D Segmentation for Radiology applications

• MICCAI HECKTOR Workshop 2020 Oropharyngeal Tumour Segmentation Using Ensemble 3D PET-CT Fusion Networks for the HECKTOR Challenge Suraj Pai, Chinmay Rao, Ibrahim Hadzic, Ivan Zhovannik, Dennis Bontempi, Andre Dekker, Jonas Teuwen, Alberto Traverso

Open-science through datasets, challenges and tools.

• arxiv Generating Synthetic Computed Tomography for Radiotherapy: SynthRAD2023 Challenge Report Evi M. C. Huijben, Maarten L. Terpstra, Arthur Jr. Galapon, Suraj Pai,…, Zoltan Perko, Matteo Maspero

💬 Invited Talks

• Invited Talk at Novartis Biomedical Institutes of Research
• Invited Talk at DL IndabaX Morocco Conference 2024
• Invited Guest on The Human Condition Podcast

👍🏼 Miscellaneous

• Reviewer for npj Breast Cancer, Nature Scientific Reports, JOSS, MICCAI
• Awarded Brigham Research Institute Microgrant