Making the Subjective Objective: Machine Learning and Rhinoplasty

Abstract

Background Machine learning represents a new frontier in surgical innovation. The ranking Convolutional Neural Network (CNN) is a novel machine learning algorithm that helps elucidate patterns and features of aging that are not always appreciable with the human eye. Objectives The authors sought to determine the impact of aesthetic rhinoplasty on facial aging employing a multidimensional facial recognition and comparison software. Methods A retrospective chart review and subsequent analysis was carried out on all female patients who underwent open rhinoplasty with the senior author from 2014 through 2018 and had postoperative photos at 12 or more weeks follow-up. All photos were analyzed with Microsoft Azure Face API (Redmond, WA), which estimates patients’ age by cropping the face from a photograph and then extracting a CNN-based prediction through multiple deep neural networks. Results A total of 100 patients ultimately met full inclusion criteria. The average post-surgical follow up for this cohort was 29 weeks (median, 14 weeks; range, 12-64 weeks). Patients ranged from 16 to 72 years old (mean, 32.75 years; median, 28.00 years; standard deviation, 12.79 years). The ranking CNN algorithm on average estimated patients preoperatively to be 0.03 years older than their actual age. The correlation coefficient between actual age and predicted preoperative age was r = 0.91. On average, patients were found to look younger post-open rhinoplasty (−3.10 vs 0.03 years, P < 0.0001). Conclusions The ranking CNN algorithm is both accurate and precise in estimating human age before and after cosmetic rhinoplasty. Given the resulting data, the effects of open rhinoplasty on reversing signs of facial aging should be revisited. Level of Evidence: 4

A Systematic Review of Artificial Intelligence Applications in Plastic Surgery: Looking to the Future

“…2 ). 27 AI models were frequently burdened by phase 0 limitations related to data, statistical performance, and workflow (n = 26) or phase 1 limitations related to technical performance and safety (n = 28). However, many studies were limited by multiple factors from various limitation phases (mean limitations per study = 3.11, n = 44).…”

Perceived Age and Attractiveness Using Facial Recognition Software in Rhinoplasty Patients: A Proof-of-Concept Study

“…27,28 Diminished skin elasticity, atrophy of support structures, and ossification of the fibrocartilaginous framework may contribute to changes in nasal morphology associated with aging, which can be addressed through rhinoplasty. 27,28 Through implementing a Convolutional Neural Network algorithm, Dorfman et al 29 similarly established that aesthetic rhinoplasty conferred reductions in perceived age in an exclusive cohort of female patients. However, it is important to note that this study had a relatively short follow-up period (29 weeks), and thus, postoperative swelling could have impacted conclusions about the role of rhinoplasty on perceived age.…”

Machine Learning Demonstrates High Accuracy for Disease Diagnosis and Prognosis in Plastic Surgery

“… 18 , 37 , 39 , 40 , 46 , 51 , 53 , 59 , 63 Further, in conditions in which there are well-established correlations between certain risk markers and an outcome of interest, such as deranged blood tests on admission and AKI in burn patients, ML yielded highly accurate predictive algorithms. 38 , 44 , 55 24 47 However, attempts to include weakly related risk markers resulted in algorithms that had an overall lower predictive accuracy, rendering them unsafe for clinical practice. This review further identified that some plastic surgery subspecialties, such as hand surgery, have yet to incorporate this technology.…”