Approved Research

Multimodal biomedical AI foundation model and its downstream meta-analysis studies

Lay summary

Aim: Our research project aims to harness the power of artificial intelligence (AI) to revolutionize healthcare and improve our understanding of complex diseases. We will develop a cutting-edge Multimodal Biomedical AI Foundation Model, capable of analyzing diverse types of medical data, including images, doctor's notes, and genetic information.

Scientific Rationale:  Based on the latest advancements in large language models like GPT-3.5 or LLaMA, we will build a multimodal foundation modal on a massive amount of data, which can understand the complex relationships between different types of medical information. This will enable us to better understand how genetics, medical images, and patient notes contribute to the development and progression of diseases.

Project Duration: Our research project will take place over a three-year period, during which we will work on building and refining the AI model using advanced techniques.

Public Health Impact: By creating a Multimodal Biomedical AI Foundation Model, we aim to improve healthcare in several ways. Firstly, the model will be open-sourced and help researchers in this field to further analyse the broad range of medical data. Secondly, the AI model's ability to analyze genetics and medical images together could reveal new insights into the causes of diseases, potentially leading to the development of innovative treatments and preventive strategies. Ultimately, our research seeks to enhance the way we understand and manage complex diseases, bringing us one step closer to a healthier and happier population.