A machine learning technique has been found non-inferior to the traditional immunohistochemistry in predicting molecular biomarker expression. The pathological review of tumour samples, even for common molecular biomarkers such as estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2), is time consuming. Moreover, there is not always concordance between pathologists on the interpretation of samples.
Artificial intelligence (AI) and machine learning technologies are being applied to address this variation as well as to improve reliability and add efficiency. Currently, such technology can differentiate between cancerous and non-cancerous tissue as well as determine presence of metastases in lymph nodes and perform tumour grading.
Now, investigators have conducted a retrospective, single-institution study to test the ability of a machine learning technique — referred to as morphological-based molecular profiling — to assess hormonal status of more than 20,000 digitised hematoxylin-eosin (H&E) pathology specimens from a microarray library of more than 5000 breast cancer patients.
AI and machine learning are emerging technologies that are finding a role in many aspects of modern life, including medicine. These data represent a first effort to utilise a training set to gauge utility for pathological review of samples.