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

Age Reduction After Face-Lift Surgery in Chinese Population: An Outcome Study Using Artificial Intelligence and Objective Observer-Based Assessment

BACKGROUND

The literature is replete with favorable face-lift results, yet the objective facial rejuvenation outcome measures in Chinese women have remained poorly understood.

OBJECTIVE

The purpose of the study is to objectively evaluate the apparent age (AA) reduction in Chinese women following face-lift by artificial intelligence (AI) and objective observers.

METHODS

Standardized pre- and postoperative (1-year) images of 48 patients undergoing face-lift procedures were analyzed by AI to estimate AA. Additionally, 10 blinded, naive observers viewed each patient's images and assessed AA. The accuracy of AA and reduction in AA were evaluated and compared between the two methods. FACE-Q surveys were employed to measure patient-reported facial esthetic outcomes.

RESULTS

The AI demonstrated higher precision than the observers in age estimation, with a mean absolute error of 3.34 years and 90% Pearson correlation. AA reduction generated by AI was significantly lower than that by observers, with a mean reduction of 3.75 ± 3.93 and 4.51 ± 1.20, respectively (p < 0.05). However, both methods showed less AA reduction than patient self-appraisal (- 7.3 years). Improvements in facial rejuvenation following face-lift surgery is relevant to the patient's preoperative aging status. Patients whose pre-AA was greater than chronological age (CA) became "back to normal," while those whose pre-AA was less than CA became "turning back the clock."

CONCLUSION

The utilization of AI could provide objective, evidence-based data in the field of face-lift surgery. As a simple, complete, and time-sparing method, AI is expected to be routinely used in clinical trials and practice.

LEVEL OF EVIDENCE IV

This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Prediction of xerostomia in elderly based on clinical characteristics and salivary flow rate with machine learning

Xerostomia may be accompanied by changes in salivary flow rate and the incidence increases in elderly. We aimed to use machine learning algorithms, to identify significant predictors for the presence of xerostomia. This study is the first to predict xerostomia with salivary flow rate in elderly based on artificial intelligence. In a cross-sectional study, 829 patients with oral discomfort were enrolled, and six features (sex, age, unstimulated and stimulated salivary flow rates (UFR and SFR, respectively), number of systemic diseases, and medication usage) were used in four machine learning algorithms to predict the presence of xerostomia. The incidence of xerostomia increased with age. The SFR was significantly higher than the UFR, and the UFR and SFR were significantly correlated. The UFR, but not SFR, decreased with age significantly. In patients more than 60 years of age, the UFR had a significantly higher predictive accuracy for xerostomia than the SFR. Using machine learning algorithms with tenfold cross-validation, the prediction accuracy increased significantly. In particular, the prediction accuracy of the multilayer perceptron (MLP) algorithm that combined UFR and SFR data was significantly better than either UFR or SFR individually. Moreover, when sex, age, number of systemic diseases, and number of medications were added to the MLP model, the prediction accuracy increased from 56 to 68%.

Automatic labeling of facial zones for digital clinical application: An ensemble of semantic segmentation models

INTRODUCTION

The application of artificial intelligence to facial aesthetics has been limited by the inability to discern facial zones of interest, as defined by complex facial musculature and underlying structures. Although semantic segmentation models (SSMs) could potentially overcome this limitation, existing facial SSMs distinguish only three to nine facial zones of interest.

METHODS

We developed a new supervised SSM, trained on 669 high-resolution clinical-grade facial images; a subset of these images was used in an iterative process between facial aesthetics experts and manual annotators that defined and labeled 33 facial zones of interest.

RESULTS

Because some zones overlap, some pixels are included in multiple zones, violating the one-to-one relationship between a given pixel and a specific class (zone) required for SSMs. The full facial zone model was therefore used to create three sub-models, each with completely non-overlapping zones, generating three outputs for each input image that can be treated as standalone models. For each facial zone, the output demonstrating the best Intersection Over Union (IOU) value was selected as the winning prediction.

CONCLUSIONS

The new SSM demonstrates mean IOU values superior to manual annotation and landmark analyses, and it is more robust than landmark methods in handling variances in facial shape and structure.

Autonomous Robotic Surgery: Has the Future Arrived?

Autonomous robotic surgery represents a pioneering field dedicated to the integration of robotic systems with varying degrees of autonomy for the execution of surgical procedures. This paradigm shift is made possible by the progressive integration of artificial intelligence (AI) and machine learning (ML) into the realm of surgical interventions. While the majority of autonomous robotic systems remain in the experimental phase, a notable subset has successfully transitioned into clinical applications. Noteworthy procedures, such as venipuncture, hair implantations, intestinal anastomosis, total knee replacement, cochlear implant, radiosurgery, and knot tying, among others, exemplify the current capabilities of autonomous surgical systems. This review endeavors to comprehensively address facets of autonomous robotic surgery, commencing with a concise elucidation of fundamental concepts and traversing the pivotal milestones in the historical evolution of robotic surgery. This historical trajectory underscores the incremental assimilation of autonomous systems into surgical practices. This review aims to address topics related to autonomous robotic surgery, starting with a description of fundamental concepts and going through the milestones in robotic surgery history that also show the gradual incorporations of autonomous systems. It also includes a discussion of the key benefits and risks of this technology, the degrees of autonomy in surgical robots, their limitations, the current legal regulations governing their usage, and the main ethical concerns inherent to their nature.

A Narrative Review of Artificial Intelligence (AI) for Objective Assessment of Aesthetic Endpoints in Plastic Surgery

Notoriously characterized by subjectivity and lack of solid scientific validation, reporting aesthetic outcome in plastic surgery is usually based on ill-defined end points and subjective measures very often from the patients' and/or providers' perspective. With the tremendous increase in demand for all types of aesthetic procedures, there is an urgent need for better understanding of aesthetics and beauty in addition to reliable and objective outcome measures to quantitate what is perceived as beautiful and attractive. In an era of evidence-based medicine, recognition of the importance of science with evidence-based approach to aesthetic surgery is long overdue. View the many limitations of conventional outcome evaluation tools of aesthetic interventions, objective outcome analysis provided by tools described to be reliable is being investigated such as advanced artificial intelligence (AI). The current review is intended to analyze available evidence regarding advantages as well as limitations of this technology in objectively documenting outcome of aesthetic interventions. It has shown that some AI applications such as facial emotions recognition systems are capable of objectively measuring and quantitating patients' reported outcomes and defining aesthetic interventions success from the patients' perspective. Though not reported yet, observers' satisfaction with the results and their appreciation of aesthetic attributes may also be measured in the same manner.Level of Evidence III This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

"Facekit"-Toward an Automated Facial Analysis App Using a Machine Learning-Derived Facial Recognition Algorithm

Introduction: Multiple tools have been developed for facial feature measurements and analysis using facial recognition machine learning techniques. However, several challenges remain before these will be useful in the clinical context for reconstructive and aesthetic plastic surgery. Smartphone-based applications utilizing open-access machine learning tools can be rapidly developed, deployed, and tested for use in clinical settings. This research compares a smartphone-based facial recognition algorithm to direct and digital measurement performance for use in facial analysis. Methods: Facekit is a camera application developed for Android that utilizes ML Kit, an open-access computer vision Application Programing Interface developed by Google. Using the facial landmark module, we measured 4 facial proportions in 15 healthy subjects and compared them to direct surface and digital measurements using intraclass correlation (ICC) and Pearson correlation. Results: Measurement of the naso-facial proportion achieved the highest ICC of 0.321, where ICC > 0.75 is considered an excellent agreement between methods. Repeated measures analysis of variance of proportion measurements between ML Kit, direct and digital methods, were significantly different (F[2,14] = 6-26, P<<.05). Facekit measurements of orbital, orbitonasal, naso-oral, and naso-facial ratios had overall low correlation and agreement to both direct and digital measurements (R<<0.5, ICC<<0.75). Conclusion: Facekit is a smartphone camera application for rapid facial feature analysis. Agreement between Facekit's machine learning measurements and direct and digital measurements was low. We conclude that the chosen pretrained facial recognition software is not accurate enough for conducting a clinically useful facial analysis. Custom models trained on accurate and clinically relevant landmarks may provide better performance.

Applications of Machine Learning in Facial Cosmetic Surgeries: A Scoping Review

OBJECTIVE

To review the application of machine learning (ML) in the facial cosmetic surgeries and procedures METHODS AND MATERIALS: Electronic search was conducted in PubMed, Scopus, Embase, Web of Science, ArXiv and Cochrane databases for the studies published until August 2022. Studies that reported the application of ML in various fields of facial cosmetic surgeries were included. The studies' risk of bias (ROB) was assessed using the QUADAS-2 tool and NIH tool for before and after studies.

RESULTS

From 848 studies, a total of 29 studies were included and categorized in five groups based on the aim of the studies: outcome evaluation (n = 8), face recognition (n = 7), outcome prediction (n = 7), patient concern evaluation (n = 4) and diagnosis (n = 3). Total of 16 studies used public data sets. ROB assessment using QUADAS-2 tool revealed that six studies were at low ROB, five studies were at high ROB, and others had moderate ROB. All studies assessed with NIH tool showed fair quality. In general, all studies showed that using ML in the facial cosmetic surgeries is accurate enough to benefit both surgeons and patients.

CONCLUSION

Using ML in the field of facial cosmetic surgery is a novel method and needs further studies, especially in the fields of diagnosis and treatment planning. Due to the small number of articles and the qualitative analysis conducted, we cannot draw a general conclusion about the impact of ML in the sphere of facial cosmetic surgery.

LEVEL OF EVIDENCE IV

This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Bard Versus the 2022 American Society of Plastic Surgeons In-Service Examination: Performance on the Examination in Its Intern Year

Background

Bard is a conversational generative artificial intelligence (AI) platform released by Google (Mountain View, CA) to the public in May 2023.

Objectives

This study investigates the performance of Bard on the American Society of Plastic Surgeons (ASPS) In-Service Examination to compare it to residents' performance nationally. We hypothesized that Bard would perform best on the comprehensive and core surgical principles portions of the examination.

Methods

Google's 2023 Bard was used to answer questions from the 2022 ASPS In-Service Examination. Each question was asked as written with the stem and multiple-choice options. The 2022 ASPS Norm Table was utilized to compare Bard's performance to that of subgroups of plastic surgery residents.

Results

A total of 231 questions were included. Bard answered 143 questions correctly corresponding to an accuracy of 62%. The highest-performing section was the comprehensive portion (73%). When compared with integrated residents nationally, Bard scored in the 74th percentile for post-graduate year (PGY)-1, 34th percentile for PGY-2, 20th percentile for PGY-3, 8th percentile for PGY-4, 1st percentile for PGY-5, and 2nd percentile for PGY-6.

Conclusions

Bard outperformed more than half of the first-year integrated residents (74th percentile). Its best sections were the comprehensive and core surgical principle portions of the examination. Further analysis of the chatbot's incorrect questions might help improve the overall quality of the examination's questions.

Quantization of extraoral free flap monitoring for venous congestion with deep learning integrated iOS applications on smartphones: a diagnostic study

BACKGROUND

Free flap monitoring is essential for postmicrosurgical management and outcomes but traditionally relies on human observers; the process is subjective and qualitative and imposes a heavy burden on staffing. To scientifically monitor and quantify the condition of free flaps in a clinical scenario, we developed and validated a successful clinical transitional deep learning (DL) model integrated application.

MATERIAL AND METHODS

Patients from a single microsurgical intensive care unit between 1 April 2021 and 31 March 2022, were retrospectively analyzed for DL model development, validation, clinical transition, and quantification of free flap monitoring. An iOS application that predicted the probability of flap congestion based on computer vision was developed. The application calculated probability distribution that indicates the flap congestion risks. Accuracy, discrimination, and calibration tests were assessed for model performance evaluations.

RESULTS

From a total of 1761 photographs of 642 patients, 122 patients were included during the clinical application period. Development (photographs =328), external validation (photographs =512), and clinical application (photographs =921) cohorts were assigned to corresponding time periods. The performance measurements of the DL model indicate a 92.2% training and a 92.3% validation accuracy. The discrimination (area under the receiver operating characteristic curve) was 0.99 (95% CI: 0.98-1.0) during internal validation and 0.98 (95% CI: 0.97-0.99) under external validation. Among clinical application periods, the application demonstrates 95.3% accuracy, 95.2% sensitivity, and 95.3% specificity. The probabilities of flap congestion were significantly higher in the congested group than in the normal group (78.3 (17.1)% versus 13.2 (18.1)%; 0.8%; 95% CI, P <0.001).

CONCLUSION

The DL integrated smartphone application can accurately reflect and quantify flap condition; it is a convenient, accurate, and economical device that can improve patient safety and management and assist in monitoring flap physiology.

Machine Learning and Artificial Intelligence in Surgical Research

Machine learning, a subtype of artificial intelligence, is an emerging field of surgical research dedicated to predictive modeling. From its inception, machine learning has been of interest in medical and surgical research. Built on traditional research metrics for optimal success, avenues of research include diagnostics, prognosis, operative timing, and surgical education, in a variety of surgical subspecialties. Machine learning represents an exciting and developing future in the world of surgical research that will not only allow for more personalized and comprehensive medical care.

A Surgeon's Guide to Artificial Intelligence-Driven Predictive Models

Artificial intelligence (AI) focuses on processing and interpreting complex information as well as identifying relationships and patterns among complex data. Artificial intelligence- and machine learning (ML)-driven predictions have shown promising potential in influencing real-time decisions and improving surgical outcomes by facilitating screening, diagnosis, risk assessment, preoperative planning, and shared decision-making. Fundamental understanding of the algorithms, as well as their development and interpretation, is essential for the evolution of AI in surgery. In this article, we provide surgeons with a fundamental understanding of AI-driven predictive models through an overview of common ML and deep learning algorithms, model development, performance metrics and interpretation. This would serve as a basis for understanding ML-based research, while fostering new ideas and innovations for furthering the reach of this emerging discipline.

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

Surgical complications pose significant challenges for surgeons, patients, and health care systems as they may result in patient distress, suboptimal outcomes, and higher health care costs. Artificial intelligence (AI)-driven models have revolutionized the field of surgery by accurately identifying patients at high risk of developing surgical complications and by overcoming several limitations associated with traditional statistics-based risk calculators. This article aims to provide an overview of AI in predicting surgical complications using common machine learning and deep learning algorithms and illustrates how this can be utilized to risk stratify patients preoperatively. This can form the basis for discussions on informed consent based on individualized patient factors in the future.

Applying Machine Learning to Determine Popular Patient Questions About Mentoplasty on Social Media

PURPOSE

Patient satisfaction in esthetic surgery often necessitates synergy between patient and physician goals. The authors aim to characterize patient questions before and after mentoplasty to reflect the patient perspective and enhance the physician-patient relationship.

METHODS

Mentoplasty reviews were gathered from Realself.com using an automated web crawler. Questions were defined as preoperative or postoperative. Each question was reviewed and characterized by the authors into general categories to best reflect the overall theme of the question. A machine learning approach was utilized to create a list of the most common patient questions, asked both preoperatively and postoperatively.

RESULTS

A total of 2,012 questions were collected. Of these, 1,708 (84.9%) and 304 (15.1%) preoperative and postoperative questions, respectively. The primary category for patients preoperatively was "eligibility for surgery" (86.3%), followed by "surgical techniques and logistics" (5.4%) and "cost" (5.4%). Of the postoperative questions, the most common questions were about "options to revise surgery" (44.1%), "symptoms after surgery" (27.0%), and "appearance" (26.3%). Our machine learning approach generated the 10 most common pre- and postoperative questions about mentoplasty. The majority of preoperative questions dealt with potential surgical indications, while most postoperative questions principally addressed appearance.

CONCLUSIONS

The majority of mentoplasty patient questions were preoperative and asked about eligibility of surgery. Our study also found a significant proportion of postoperative questions inquired about revision, suggesting a small but nontrivial subset of patients highly dissatisfied with their results. Our 10 most common preoperative and postoperative question handout can help better inform physicians about the patient perspective on mentoplasty throughout their surgical course. Level of Evidence V This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Toward a Universal Measure of Facial Difference Using Two Novel Machine Learning Models

A sensitive, objective, and universally accepted method of measuring facial deformity does not currently exist. Two distinct machine learning methods are described here that produce numerical scores reflecting the level of deformity of a wide variety of facial conditions.

METHODS

The first proposed technique utilizes an object detector based on a cascade function of Haar features. The model was trained using a dataset of 200,000 normal faces, as well as a collection of images devoid of faces. With the model trained to detect normal faces, the face detector confidence score was shown to function as a reliable gauge of facial abnormality. The second technique developed is based on a deep learning architecture of a convolutional autoencoder trained with the same rich dataset of normal faces. Because the convolutional autoencoder regenerates images disposed toward their training dataset (ie, normal faces), we utilized its reconstruction error as an indicator of facial abnormality. Scores generated by both methods were compared with human ratings obtained using a survey of 80 subjects evaluating 60 images depicting a range of facial deformities [rating from 1 (abnormal) to 7 (normal)].

RESULTS

The machine scores were highly correlated to the average human score, with overall Pearson's correlation coefficient exceeding 0.96 (P < 0.00001). Both methods were computationally efficient, reporting results within 3 seconds.

CONCLUSIONS

These models show promise for adaptation into a clinically accessible handheld tool. It is anticipated that ongoing development of this technology will facilitate multicenter collaboration and comparison of outcomes between conditions, techniques, operators, and institutions.

Artificial intelligence, machine learning, and deep learning for clinical outcome prediction

AI is a broad concept, grouping initiatives that use a computer to perform tasks that would usually require a human to complete. AI methods are well suited to predict clinical outcomes. In practice, AI methods can be thought of as functions that learn the outcomes accompanying standardized input data to produce accurate outcome predictions when trialed with new data. Current methods for cleaning, creating, accessing, extracting, augmenting, and representing data for training AI clinical prediction models are well defined. The use of AI to predict clinical outcomes is a dynamic and rapidly evolving arena, with new methods and applications emerging. Extraction or accession of electronic health care records and combining these with patient genetic data is an area of present attention, with tremendous potential for future growth. Machine learning approaches, including decision tree methods of Random Forest and XGBoost, and deep learning techniques including deep multi-layer and recurrent neural networks, afford unique capabilities to accurately create predictions from high dimensional, multimodal data. Furthermore, AI methods are increasing our ability to accurately predict clinical outcomes that previously were difficult to model, including time-dependent and multi-class outcomes. Barriers to robust AI-based clinical outcome model deployment include changing AI product development interfaces, the specificity of regulation requirements, and limitations in ensuring model interpretability, generalizability, and adaptability over time.