Breast cancer classed as HER2-positive – defined as an immunohistochemical (IHC) score of three or more – accounts for somewhere between 20% to 25% of all patients with the disease.
According to Lunit, patients identified by the AI model as having a high proportion of HER2 IHC 3+ tumour cells exhibited better clinical outcomes than those identified through traditional HER2 ...
Prediction of HER2 overexpression status and IHC score as seen in "Predicting the HER2 status in esophageal cancer from tissue microarrays using convolutional neural networks".