Improve the reliability and performance of Vision based Machine Learning models to provide more valuable insights to researchers.

Biomedical research, particularly preclinical research, is a complex and challenging field with a high failure rate of 98% in pharmaceutical research investment. Extracting relevant information from preclinical research papers involves synthesizing information from various sources, which is a demanding task that requires domain-specific knowledge. Natural language processing, specifically Large Language Models (LLMs), has demonstrated tremendous potential in extracting information from unstructured text. This project aims to train and deploy LLMs to accurately extract biological entities and other relevant information from preclinical research data. The extracted entities will be used to create ontologies and a knowledge base to facilitate the discovery of correlations and evidence that could hasten drug discovery. By providing a comprehensive and structured representation of the extracted data, this approach could help researchers develop new insights, and potentially accelerate the development of new treatments for various diseases.

Faculty Supervisor:

Nick Koudas

Student:

Partner:

BenchSci

Discipline:

Computer science

Sector:

Artificial Intelligence; Technology; Health and Related Sciences & Technology

University:

University of Toronto

Program: