The power of artificial intelligence (AI) applied to endometrial carcinoma microscopy images is demonstrated in a new study. The research team provides novel insights that could improve uterine cancer diagnosis and treatment.
Researchers used AI to analyze microscopy images of thousands of endometrial carcinoma patients. This model was not a “black box,” but the researchers could show which image features were relevant for its predictions by reverse-engineering it. The AI model provided the team with important novel insights that could improve endometrial cancer diagnostics, prognosis, and management in the future.
Researchers studied the morphology of tumors with the same molecular alteration to better understand the effect these changes have on the appearance of the tumor. The AI model has directed them to important areas within and outside the tumor.
The number of variables (molecular, tumor morphology, patient data) in cancer diagnostics has grown exponentially, complicating patient prognosis prediction. AI predictions can teach pathologists in return by identifying novel morphological details on microscopy slide images with predictive value, for example, by training unbiased AI models.
Related Content: Choroideremia – Novel Imaging Approach