Fusing Multi-Modal Medical Data with Reliable AI
Postdoctoral Research Associate at Imperial College London. I develop trustworthy AI frameworks - focusing on uncertainty quantification and multi-modal fusion - to ensure state-of-the-art deep learning models are reliable, explainable, and safe.
Multi-Modal Fusion
Synthesising complementary information from medical imaging (MICCAI/ISBI) and clinical records. I develop fusion architectures that robustly integrate diverse data modalities.
Uncertainty Quantification
Designing deep learning frameworks that rigorously quantify uncertainty. My work ensures AI systems are aware of their own limitations, crucial for safe clinical deployment.
Medical Foundation Models
Adapting state-of-the-art Large Language Models (LLMs) for healthcare. I focus on fine-tuning foundation models to be both explainable and clinically accurate.
Recent News
Featured Research
Deep Evidential Fusion for Multimodal Medical Image Segmentation
My 2024 paper in Information Fusion introduces a deep evidential fusion framework for multimodal medical image segmentation. It addresses the “black box” problem by explicitly quantifying uncertainty and learning the reliability of each imaging source — shifting the emphasis from raw accuracy to clinical trustworthiness.
Dive DeeperSelected Publications
Ling Huang (first author), Yucheng Xing, Qika Lin, Su Ruan, Mengling Feng
Yucheng Xing, Ling Huang (corresponding author), Jingying Ma, Ruping Hong, Jiangdong Qiu, Pei Liu, Kai He, Huazhu Fu, Mengling Feng
