Present and future of machine learning in breast surgery: systematic review

Area of Science:

• Medical Informatics
• Computational Biology
• Surgical Technology

Background:

• Machine learning (ML) offers powerful tools for pattern detection and decision-making in complex data.
• The application of ML in breast surgery is rapidly evolving, necessitating awareness among breast surgeons.
• Understanding ML's potential is crucial for advancing patient care in breast surgery.

Purpose of the Study:

• To systematically review the applications of machine learning (ML) and artificial intelligence (AI) in breast surgery.
• To identify key areas where ML is being utilized within the field of breast surgery.
• To evaluate the efficacy and potential of ML in improving surgical outcomes and patient care.

Main Methods:

• A systematic literature search was performed across Embase, MEDLINE, Cochrane database, and Google Scholar.
• The search covered original articles from inception to December 2021 focusing on ML/AI in breast surgery.
• Inclusion criteria led to the selection of 14 studies involving 73,847 patients.

Main Results:

• Four primary ML application areas were identified: predictive modeling, breast imaging analysis, patient screening/triaging, and network utility for detection.
• ML demonstrated significant value in preoperative planning, surgical information provision, and anatomical visualization.
• ML consistently outperformed traditional statistical models in predicting mortality, morbidity, and quality of life outcomes.

Conclusions:

• Machine learning (ML) shows considerable promise for enhancing breast surgery outcomes and patient-centered care.
• ML applications can support surgical planning, visualization, and navigation.
• Significant limitations and ethical considerations must be addressed for the integration of AI into surgical practice.