Welcome to the Rasool Lab at Moffitt Cancer Center & Research Institute

Building Trustworthy AI for Cancer Research and Care


The Rasool Lab at Moffitt Cancer Center & Research Institute develops advanced machine learning methods to transform cancer research and clinical care. We specialize in creating trustworthy, multimodal AI systems that bring together radiology, pathology, molecular, electronic health records (EHRs), and cancer registry data to improve patient outcomes.

 

Our Core Research Areas

We focus on five interconnected themes that drive our mission:

  1. Multimodal Learning — Integrating imaging, molecular, text, and clinical data to discover new biomarkers and predictive models for cancer
  2. Federated Learning — Developing privacy-preserving frameworks that enable collaboration across institutions while protecting patient data
  3. Large Language Models (LLMs) in Radiology & Pathology Informatics — Applying LLMs to enhance clinical reporting, streamline pathology data abstraction, and accelerate cancer registry workflows
  4. Patient-Facing Chatbots — Studying and designing safe, reliable conversational AI tools that empower patients with accurate, accessible cancer information
  5. Early Detection of Cancer and Cancer-Related Cachexia — Creating AI-driven biomarkers and predictive models to identify disease earlier and improve interventions

 

Impact

Our work is supported by federal and state grants, collaborative team science, and clinical partnerships, with recognition from organizations such as AACR, RSNA, and ASTRO. By uniting diverse data modalities with cutting-edge AI, we aim to deliver innovations that are scientifically rigorous, clinically relevant, and patient-centered. We are committed to collaboration, innovation, and mentorship, training the next generation of researchers at the intersection of engineering, computer science, and oncology.

                                                                      

                                          Linkedin                Github            HuggingFace        Google Scholar