Artificial intelligence is changing the face of plastic surgery

Artificial intelligence (AI) has taken the world by storm, sticking finger in various sectors. In healthcare delivery systems, AI has been showing promising results and has therefore been an important area of research. The rapid progression of technology has allowed the growth of many processes which were earlier reliant on manpower. Artificial intelligence has become popular in the field of plastic surgery in a number of settings and procedures. The current applications of AI in plastic surgery lies in aesthetic surgeries, burn care, craniofacial surgery, and more. 

Artificial intelligence is a transformative technology that is being implemented in various branches of medicine. While the complete potential of the technology is yet to be realized, researchers are focusing on exploring ways to use machine learning and other subsets of AI for dramatic improvement of patient outcomes. The area of machine learning aims to understand and process the large amounts of data, which makes it an important tool for handling the waves of electronic data in healthcare industry. 

As plastic surgery isn’t only about the improvement of aesthetics or cosmetic effects, but also about reconstructing and forming new contours. Plastic surgery is also performed to restore the form and function of the part of the body. This includes managing burns, treating skin cancer, correction of congenital defects such as cleft lips and cleft palette, and reconstruction after accidents and surgeries. AI can be employed to enhance the results of these procedures. 

The sub-disciplines in AI are:

Machine learning: In this, algorithms are designed to uncover the associations in large data sets using pattern recognition in interacting variables. This is further categorized into supervised and unsupervised learning.

Supervised learning involves the application that is tested with photographs to monitor the post-operative viability of free flap based on skin color.

Unsupervised learning is the interpretation of large amounts of unlabeled genetic data without any training set.

Deep learning: These learning models use artificial neural networks to improve the performance in terms of prediction and accuracy with continued training. For example, a deep learning convolutional network can be used to determine status of rhinoplasty via photographs. Its application is also capable of identifying melanoma (skin cancer) in the images of biopsied lesions taken with a smartphone.

Natural language processing: This is a machine learning software which enables understanding, interpreting, and manipulating of human language. A bot within a smartphone app can be used to provide answers to frequently asked questions of patients during the preoperative phase.

Facial recognition: AI software have been designed to recognize human faces by using biometrics to map facial features. This data is compared with a database of photographs. Facial recognition of neural networks also enables gender-typing of transgender women after facial feminization surgery.

As imaging and videos are already part of the plastic surgeon’s planning, use of robotics and advanced software for designs and operation. People who are looking to travel overseas for a cosmetic procedure may consider locations that offer advanced surgical procedures. Plastic surgery in Turkey is one such option where the trained plastic surgeon employ the latest imaging and robotic techniques for surgery.

Use of AI in plastic surgery may include, but is not limited to:

Aesthetic surgery: Cosmetic procedure for the face requires careful pre-operative planning which includes taking measurement of the patient’s facial dimensions. Beauty is subjective and a patient must have realistic expectation from the outcomes of the surgery. However, many people share a general feeling about facial beauty, as per the studies. AI can assist surgeons in planning the surgeries as well as guide patients’ to make a sound choice using large datasets of facial images.

AI uses a comprehensive facial image database to assess the features and dimensions of the face. Combining the information from the database with measurements of a patient’s unique facial ratio, AI presents specific information to the surgeon about the best possible course of action regarding the surgery. This process involves the use of indicators such as skin color, evenness of skin and facial symmetry to ensure an objective assessment. The system also provides a real-time simulation which enables the patient to see what kind of look they will have after the procedure.

Burn care

AI is used to accurately assess the total surface area of a burn which plays an essential role in delivering appropriate treatment. The system also aids in predicting whether a burn wound can heal without any surgery. Researchers have reported the use of reflectance spectrometry and an artificial neural network for prediction of the time of healing of a burn wound - whether it would take more or less than 14 days to heal.

Craniofacial surgery

This is a type of surgery in which the bones, muscles, and skin of the skull are moved. A congenital condition (birth defect) known as craniosynostosis causes a disorder of bone growth. It is caused by a premature fusion of the skull, which creates an abnormal shape of the skull and also leads to increased pressure on the brain. It can also lead to delayed growth. Infants as young as two months old may need to have a major craniofacial surgery for re-shaping the skull. The detection and treatment of the condition sooner is better. Normally, images of babies’ heads and CT scans are used to examine their skull shape and plan the surgery. An AI trained to classify the shape of a baby’s skull  to better diagnose craniosynostosis at early stage has been introduced. This technology may also help in screening children for the condition and reduces the exposure to rays in X-rays or CT scans needed for the diagnosis.