Artificial Intelligence and Machine Learning in Surgery

Artificial intelligence (AI) as a scientific discipline has grown significantly over the past few decades. Consumer applications such as voice recognition through Siri/Google voice on Apple/Android systems, respectively, reflect basic applications of AI in daily living. Over the past few years, the principles of AI have developed exponentially and have laid down the foundation for advanced applications in the technology field, including self-driving cars, targeted advertising on social media platforms, online translation software, and so forth.

In health care, AI has made early in-roads into the fields of pathology, radiology, and dermatology, where large databases of images have been available for comparison and for training AI models. The overall concept being novel, adoption into clinical practice on a large scale has remained elusive, and much of the incorporation of AI into clinical practice has been in the research context and limited to large academic institutions.

Surgery is a dynamic field based on movement, and the surgical robot is one of the major milestones for AI in surgery. Robots can be programmed to perform autonomous tasks and previous studies have compared and demonstrated that such autonomous robots equipped with sound fundamental algorithms can outperform expert surgeons in certain deconstructed surgical skills in the ex vivo setting. Ultimately, the overarching goals for AI in surgery over the upcoming decades would be multifold and would impact pre-operative, intra-operative, and post-operative patient care. The major subdivisions of AI, namely, machine learning (ML), computer vision (CV), natural language processing (NLP), and deep learning (DL) in tandem, have the ability to radically change patient management in each of these phases of surgical care. ML algorithms can facilitate pre-operative patient risk stratification based on non-linear variables with greater accuracy than traditional risk calculators and can predict the likelihood of post-operative complications for any given patient. NLP has the potential to save hours of labor and manpower by extracting text from electronic records and automatically coding and billing appropriate levels of care. Additionally, dictation software which is widely adopted across hospitals in the United States utilizes an NLP algorithm as its core technology.

In the intra-operative setting, the futuristic autonomous surgical robot probably represents the apex achievement for AI in surgery. While the very notion of such technology is still in its bare infancy, DL models and reinforcement learning based on outcomes form the essentials for this upcoming innovation. AI also has massive potential in surgical training where teaching, practicing, and testing of surgical skill can be facilitated by such novel technology.

Through the eight articles that follow this introductory outline as part of our symposium on AI, we as authors wish to highlight the broad scope of this discipline in the field of surgery. The articles in sequence seek to explain AI and its applications from a clinician perspective, and thus provide readers with no prior knowledge of this subject, a solid grasp of the tenets of AI in surgery. The intended audience of this symposium includes surgeons in practice, surgical residents, and aspiring medical students who wish to mature as surgeon innovators in the years to come.

The symposium of articles that follow are as listed below:

1. A Surgeon’s Guide to Artificial Intelligence-Driven Predictive Models

2. A Synopsis of Artificial Intelligence and its Applications in Surgery.

3. Artificial Intelligence and Machine Learning in Prediction of Surgical Complications: Current State, Applications, and Implications

4. Predicting Patient-Reported Outcomes Following Surgery Using Machine Learning

5. Deep Learning Applications in Surgery: Current Uses and Future Directions

6. New Frontiers of Natural Language Processing in Surgery

7. Advancing Surgical Education: The Use of Artificial Intelligence in Surgical Training

8. Ethical, Legal, and Financial Considerations of Artificial Intelligence in Surgery

In summary, AI can truly revolutionize the care of surgical patients, teaching of surgical trainees, and has massive potential for automating processes which can ultimately culminate in reduced health care costs. As with any innovation, ethical and legal challenges will have to be overcome, and surgeons in practice will be the final effectors to embrace and utilize this technology for delivering high-quality patient care.