Abstract

Computer vision is a rapidly evolving field at the forefront of AI. Recently, outstanding progress in computer vision has led to AI systems with super-human ability to classify or detect objects in images. And recent generative AI methods have opened up new possibilities in data synthesis. This talk will provide some perspectives on how computer vision can accelerate the drug discovery process, from early stage to late stage. I will touch on research underway at Queen Mary’s Digital Environment Research Institute (DERI), for example use of computer vision in mechanism of action studies and computer vision tools for enhanced clinical trials. I will also discuss multimodal AI research where patient genotype, transcriptome and proteome can be combined with histopathological images to classify disease endotypes for drug discovery and response prediction. The talk will also provide an outlook for future research directions for personalised healthcare and in silico trials.