Discover how an AI model identifies breast tumor stages likely to progress to invasive cancer in a groundbreaking study.
- A new AI model developed by researchers from MIT and ETH Zurich can identify different stages of ductal carcinoma in situ (DCIS) from breast tissue images.
- The model analyzes cell state and organization in tissue samples to determine the stage of DCIS, helping clinicians streamline diagnosis without labor-intensive tests.
- The researchers created a large dataset of 560 tissue sample images to train the AI model, which can identify eight important markers of DCIS.
- By considering both cell proportion and arrangement in tissue samples, the AI model significantly improves accuracy in diagnosing DCIS.
- The model’s ability to analyze tissue organization could be applied to other types of cancer as well as neurodegenerative conditions, showing promise for future research and clinical applications.
Source link
Oncology, Radiology