Artificial intelligence in breast cancer histopathology

Abstract

This is a review on the use of artificial intelligence for digital breast pathology. A systematic search on PubMed was conducted, identifying 17,324 research papers related to breast cancer pathology. Following a semimanual screening, 664 papers were retrieved and pursued. The papers are grouped into six major tasks performed by pathologists—namely, molecular and hormonal analysis, grading, mitotic figure counting, ki‐67 indexing, tumour‐infiltrating lymphocyte assessment, and lymph node metastases identification. Under each task, open‐source datasets for research to build artificial intelligence (AI) tools are also listed. Many AI tools showed promise and demonstrated feasibility in the automation of routine pathology investigations. We expect continued growth of AI in this field as new algorithms mature.

Performance of externally validated machine learning models based on histopathology images for the diagnosis, classification, prognosis, or treatment outcome prediction in female breast cancer: A systematic review

“…For example, a study showing the results of the external validation of GALEN Breast, a commercially available algorithm used to detect invasive and in situ breast carcinomas, was excluded because it was published in December 2022. 87 Besides, numerous ML models not developed for diagnosis, classification, prognosis, or treatment outcome prediction (such as some recently reviewed by Chan et al), 88 even if they became commercially available and were externally validated, were excluded for not meeting the inclusion criteria (e.g., had a different purpose). Future comprehensive analyses may be facilitated with an increased availability of external validation datasets and by enhancing adherence to standardized methods and reporting protocols.…”

Computational pathology to improve biomarker testing in breast cancer: how close are we?

“…Moreover, recent lines of evidence suggest that the presence of TILs is associated with response to different neo/adjuvant therapy regimens (Burstein et al ., 2019; El Bairi et al ., 2021). Hence, TILs assessment is being gradually adopted as part of standard cancer reporting (Chan et al , 2023).…”

Annual Review Issue: Breast Pathology

“…The application of artificial intelligence (AI) in breast pathology is currently attracting a great deal of attention with a wide spectrum of potential uses. In this ARI, Dr Ronald Chan and colleagues 14 demonstrate its feasibility in automating routine pathology investigations. The authors highlighted the increasing important role to be played by AI.…”