Researchers at Michigan Tech have developed a machine learning model for detecting breast cancer from histopathology images — tissues and cells examined under microscope. The model can classify benign and malignant tumors from these images — and it can also evaluate the uncertainty in its predictions.
In their paper recently published in the journal IEEE Transactions on Medical Imaging, graduate students Ponkrshnan Thiagarajan and Pushkar Khairnar and Susanta Ghosh (ME-EM) outline their novel probabilistic machine learning model, which outperforms similar models.