Artificial Intelligence for Oral Dysplasia Grading from Histology Images

Oral cancer is the 16th most common cancer with 389,485 new cases being diagnosed globally in 2022. Patients with a suspicious lesion are referred for a biopsy where cancer or dysplasia (pre-cancer) could be diagnosed. Currently, pathologists manually examine histology images generated from biopsies to diagnose patients, but we would like to explore the use of artificial intelligence (AI) in diagnosis. Oral cancer has a low survival rate meaning it’s important that diagnosis is quick and effective, with an accurate cancer grade also determining treatment path. The student would be working with histology images to explore the classification of dysplasia grades using AI. The objectives of this project are to develop an AI system to grade histology slides of oral dysplasia, to understand what the areas of interest are within histology slides, and to evaluate the system’s generalisability across different histology datasets. Students interested in python programming and related bioinformatics would suit this project. Students will need to find their own accommodation and be expected to present their findings orally at a research day in York on 08th September 2026.