Artificial Intelligence in Surgical Research: Accomplishments and Future Directions

Area of Science:

• Surgical Informatics
• Artificial Intelligence in Medicine
• Computational Biology

Background:

• The proliferation of electronic health records and computational power facilitates AI and machine learning (ML) in healthcare.
• Algorithmic approaches are now being developed for surgical patients across all care stages (preoperative, intraoperative, postoperative).
• Surgeons need to critically evaluate the strengths and weaknesses of these ML methodologies.

Purpose of the Study:

• To investigate the current understanding of machine learning applications in surgical practice.
• To explore future directions for ML in surgery, including risk stratification, clinical data analytics, and decision support.
• To emphasize the importance of understanding ML for appropriate implementation in surgery.

Main Methods:

• A review of the existing artificial intelligence (AI) literature was conducted.
• Emphasis was placed on contemporary ML approaches relevant to the surgical domain.
• The review focused on methods applicable to surgical practice.

Main Results:

• Conventional machine learning (ML) methods and their surgical implementations are introduced.
• The study highlights the necessity of advancing beyond traditional ML approaches due to the rise of big data.
• The review covers various ML techniques pertinent to surgical applications.

Conclusions:

• The potential impact of artificial intelligence (AI) on clinical surgery and its subspecialties is substantial.
• As AI becomes more prevalent in surgical literature and practice, understanding its mechanisms and limitations is crucial.
• Informed implementation of AI in surgery requires a thorough grasp of its capabilities and shortcomings.