Machine learning has the potential to contribute to advances in plastic surgery

Posted: 09/05/2016

A recent article published in the official medical journal of the American Society of Plastic Surgeons (ASPS) has highlighted the significance of the ever-increasing volume of electronic data being collected by the healthcare system to contribute towards advancements in plastic surgery. This is called “machine learning”, a subfield of artificial intelligence which can be used to improve care and patient outcomes. 

Machine learning studies historical data to develop algorithms capable of knowledge acquisition. It has already been used with great success to process large amounts of complex data in medicine and surgery.

Dr Jonathan Kanevsky of McGill University, Montreal and his colleagues wrote: “Machine learning has the potential to become a powerful tool in plastic surgery, allowing surgeons to harness complex clinical data to help guide key clinical decision-making.” There are five areas where they believe machine learning shows promise for improving efficiency and clinical outcomes:

• Burn surgery – machine learning has already been used to predict the healing time of burns. Algorithms could also be developed to enable the prediction of the percentage of body surface area burned.
• Microsurgery – currently has been used to monitor blood perfusion of tissue flaps, based on smartphone photographs. In the future, machine learning may be used to help recommend the best reconstructive surgery.
• Craniofacial surgery – machine learning techniques have been used for automated diagnosis of infant skull growth defects. In the future, it could be used to identify known and unknown genes responsible for cleft lip and palate.
• Hand and peripheral nerve surgery – machine learning may be useful in predicting the success of tissue-engineered nerve grafts, assisting those with spinal cord injuries and predicting the outcomes of hand surgery.
• Aesthetic surgery – machine learning could be used for predicting and simulating the outcomes of aesthetic surgery and reconstructive breast surgery. 

It is also believed that machine learning could be used to improve plastic surgery training. Training is currently assessed through written and oral examinations and subjective evaluation by physicians. However, more objective methods of assessing competency are required, which may be done through machine learning. With the development of head-mounted cameras, trainees are able to record their cases and compare with previous performance. They can track their performance and develop methods of improvement.

Elise Bevan, a senior associate in the Penningtons Manches clinical negligence team, comments: “Machine learning highlights the potential of this technology for aiding and developing research and clinical practice in plastic surgery. It has already been successful in other areas of the medical profession. But it is important to remember that computer-generated algorithms cannot replace the trained human eye and there needs to be measures in place to ensure safety and clinical relevance of the results obtained by machine learning.”

Return to news headlines