View Proposal

Proposer
Heba Elshimy
Title
Zero-Shot Medical Image Classification via Vision-Language Prompt Engineering
Goal
This aims to explore if we can leverage the knowledge embedded in large-scale pre-trained models (VLMs) to achieve clinically useful performance without the costly annotation and training typically required for medical imaging tasks
Description
This project investigates the capability of Vision-Language Models (VLMs) to perform zero-shot and few-shot medical image classification through systematic prompt engineering, without task-specific fine-tuning. As foundation models increasingly demonstrate cross-domain generalisation, understanding their boundaries in specialised medical contexts becomes essential.
Resources
Datasets: CheXpert (https://stanfordmlgroup.github.io/competitions/chexpert/), ISIC Archive (https://www.isic-archive.com/)
Background
Knowledge with huggingface and open source foundational or large-scale models
Url
Difficulty Level
High
Ethical Approval
None
Number Of Students
1
Supervisor
Heba Elshimy
Keywords
vlms, ai for healthcare
Degrees
Master of Science in Artificial Intelligence
Master of Science in Computer Science for Cyber Security
Master of Science in Data Science
Master of Science in Network Security
MSc Applied Cyber Security
Back to List