Artificial Intelligence for Cancer Imaging Data Analysis

This project leverages state-of-the-art deep learning techniques to transform cancer imaging analysis for major cancers such as brain tumors, lung cancer, prostate cancer, and breast cancer. The research focuses on developing advanced AI-driven methods for lesion segmentation, radiomics, and clinical decision-making. By integrating cutting-edge Vision-Language Models (VLMs) and pretraining paradigms such as Vision-Language Pretraining (VLP), the project aims to enhance the interpretability and accuracy of diagnostic and prognostic tasks.

Leveraging multimodal learning and contrastive learning approaches, the research will explore innovative algorithms that combine imaging, radiomics, and textual data to achieve robust feature extraction and classification. This includes adopting hybrid methods like masked prediction and image-text matching to address the challenges of data scarcity and variability in medical imaging datasets.

The outcomes will include scalable AI models tailored for real-world clinical workflows, offering precise tumor characterization, radiomic feature analysis, and aiding in personalized treatment planning. By utilizing large publicly available datasets and curating diverse imaging-text datasets, the project seeks to set a new benchmark in cancer imaging AI, ultimately improving healthcare outcomes and advancing the frontiers of medical imaging and radiomics analysis.

The ideal candidate should have strong programming skills in Python and experience with deep learning frameworks like PyTorch or TensorFlow. Knowledge of medical imaging, radiomics, and AI model development is highly desirable, along with an interest in interdisciplinary research combining AI, healthcare, and imaging data.

This project is eligible for funding but is subject to eligibility criteria & funding availability.

Page expires: 13 December 2025