Speaker: Mr. Haomiao Ni
From: Pennsylvania State University
Abstract
In the past two years, transformative AI products such as ChatGPT have underscored the profound impact of AI in our daily lives. Beyond natural language and image understanding, AI is revolutionizing healthcare, offering advantages such as assisting in diagnosis and treatment, enhancing healthcare safety and reliability, and reducing the financial burden of medical care. As a pivotal technique for processing medical imaging data, medical computer vision has played a crucial role in developing healthcare AI systems. However, it comes with its own challenges including the complexity of different diseases and the difficulty of collecting large-scale annotated data.
In this talk, I will show how we address these challenges in our research work. Firstly, I will introduce our work on robust biomedical image analysis and synthesis, demonstrating the utilization of clinical prior knowledge for robust medical image analysis and synthetic augmentation with large-scale pretraining to alleviate limited annotation issues in the medical domain. Secondly, I will introduce our research on video analysis and synthesis for AI diagnosis, where I will illustrate our proposed video-based healthcare AI systems that can process infant movement videos and patient talking videos for the diagnosis of cerebral palsy and stroke. I will also discuss our proposed generative models for video synthesis and their application to healthcare AI. Finally, I will share insights into future work in this exciting field.
For more info, please follow this link.
Read More