Project Details
Description
In the last year, we have developed an AI model able to detect tumour regions from breast cancer CT scans. The model can merge patient-specific clinical information (e.g., age, and tumour grade), cancer cells data (transcriptomics) and images (CT-Scans) to automatically identify the tumour regions.
Currently, the model can achieve high accuracy, but to make this transferable into the clinic, it is necessary to incorporate feedback from breast cancer radiologists, who would be the main users of the AI tool. This would also help us understand if the AI model can detect tumour features not easily detected by radiologists and become a support tool for therapeutic decision-making.
This project aims to bridge the gap between the AI model developed by AI researchers and radiologists who will effectively use the tool. Incorporating domain-specific expertise into the AI model will enhance its effectiveness and broaden its practical applications in clinical settings.
Currently, the model can achieve high accuracy, but to make this transferable into the clinic, it is necessary to incorporate feedback from breast cancer radiologists, who would be the main users of the AI tool. This would also help us understand if the AI model can detect tumour features not easily detected by radiologists and become a support tool for therapeutic decision-making.
This project aims to bridge the gap between the AI model developed by AI researchers and radiologists who will effectively use the tool. Incorporating domain-specific expertise into the AI model will enhance its effectiveness and broaden its practical applications in clinical settings.
Status | Finished |
---|---|
Effective start/end date | 1/04/24 → 31/07/24 |
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.