Background: The United States Medical Licensing Examination (USMLE) has been critical in medical education since 1992, testing various aspects of a medical student’s knowledge and skills through different steps, based on their training level. Artificial intelligence (AI) tools, including chatbots like ChatGPT, are emerging technologies with potential applications in medicine. However, comprehensive studies analyzing ChatGPT’s performance on USMLE Step 3 in large-scale scenarios and comparing different versions of ChatGPT are limited. Objective: This paper aimed to analyze ChatGPT’s performance on USMLE Step 3 practice test questions to better elucidate the strengths and weaknesses of AI use in medical education and deduce evidence-based strategies to counteract AI cheating. Methods: A total of 2069 USMLE Step 3 practice questions were extracted from the AMBOSS study platform. After including 229 image-based questions, a total of 1840 text-based questions were further categorized and entered into ChatGPT 3.5, while a subset of 229 questions were entered into ChatGPT 4. Responses were recorded, and the accuracy of ChatGPT answers as well as its performance in different test question categories and for different difficulty levels were compared between both versions. Results: Overall, ChatGPT 4 demonstrated a statistically significant superior performance compared to ChatGPT 3.5, achieving an accuracy of 84.7% (194/229) and 56.9% (1047/1840), respectively. A noteworthy correlation was observed between the length of test questions and the performance of ChatGPT 3.5 (ρ=–0.069; P=.003), which was absent in ChatGPT 4 (P=.87). Additionally, the difficulty of test questions, as categorized by AMBOSS hammer ratings, showed a statistically significant correlation with performance for both ChatGPT versions, with ρ=–0.289 for ChatGPT 3.5 and ρ=–0.344 for ChatGPT 4. ChatGPT 4 surpassed ChatGPT 3.5 in all levels of test question difficulty, except for the 2 highest difficulty tiers (4 and 5 hammers), where statistical significance was not reached. Conclusions: In this study, ChatGPT 4 demonstrated remarkable proficiency in taking the USMLE Step 3, with an accuracy rate of 84.7% (194/229), outshining ChatGPT 3.5 with an accuracy rate of 56.9% (1047/1840). Although ChatGPT 4 performed exceptionally, it encountered difficulties in questions requiring the application of theoretical concepts, particularly in cardiology and neurology. These insights are pivotal for the development of examination strategies that are resilient to AI and underline the promising role of AI in the realm of medical education and diagnostics.

Background: The grading process in facial palsy (FP) patients is crucial for time- and cost-effective therapy decision-making. The House-Brackmann scale (HBS) represents the most commonly used classification system in FP diagnostics. This study investigated the benefits of linking machine learning (ML) techniques with the HBS. Methods: Image datasets of 51 patients seen at the Department of Plastic, Hand, and Reconstructive Surgery at the University Hospital Regensburg, Germany, between June 2020 and May 2021, were used to build the neural network. A total of nine facial poses per patient were used to automatically determine the HBS. Results: The algorithm had an accuracy of 98%. The algorithm processed the real patient image series (i.e., nine images per patient) in 112 ms. For optimized accuracy, we found 30 training runs to be the most effective training length. Conclusion: We have developed an easy-to-use, time- and cost-efficient algorithm that provides highly accurate automated grading of FP patient images. In combination with our application, the algorithm may facilitate the FP surgeon’s clinical workflow.

Background: Reliable, time- and cost-effective, and clinician-friendly diagnostic tools are cornerstones in facial palsy (FP) patient management. Different automated FP grading systems have been developed but revealed persisting downsides such as insufficient accuracy and cost-intensive hardware. We aimed to overcome these barriers and programmed an automated grading system for FP patients utilizing the House and Brackmann scale (HBS). Methods: Image datasets of 86 patients seen at the Department of Plastic, Hand, and Reconstructive Surgery at the University Hospital Regensburg, Germany, between June 2017 and May 2021, were used to train the neural network and evaluate its accuracy. Nine facial poses per patient were analyzed by the algorithm. Results: The algorithm showed an accuracy of 100%. Oversampling did not result in altered outcomes, while the direct form displayed superior accuracy levels when compared to the modular classification form (n = 86; 100% vs. 99%). The Early Fusion technique was linked to improved accuracy outcomes in comparison to the Late Fusion and sequential method (n = 86; 100% vs. 96% vs. 97%). Conclusions: Our automated FP grading system combines high-level accuracy with cost- and time-effectiveness. Our algorithm may accelerate the grading process in FP patients and facilitate the FP surgeon’s workflow.

Background: Synkinesis of the facial musculature is a detrimental sequalae in post-paralytic facial palsy (PPFP) patients. Detailed knowledge on the technical requirements and device properties in a high-resolution ultrasound (HRUS) examination is mandatory for a reliable facial muscle assessment in PPFP patients. We therefore aimed to outline the key steps in a HRUS examination and extract an optimized workflow schema. Methods: From December 2020 to April 2021, 20 patients with unilateral synkinesis underwent HRUS. All HRUS examinations were performed by the first author using US devices with linear multifrequency transducers of 4–18 MHz, including a LOGIQ E9 and a LOGIQ S7 XDclear (GE Healthcare; Milwaukee, WI, USA), as well as Philips Affinity 50G (Philips Health Systems; Eindhoven, the Netherlands). Results: Higher-frequency and multifrequency linear probes ≥15 MHz provided superior imaging qualities. The selection of the preset program Small Parts, Breast or Thyroid was linked with a more detailed contrast of the imaging morphology of facial tissue layers. Frequency (Frq) = 15 MHz, Gain (Gn) = 25–35 db, Depth (D) = 1–1.5 cm, and Focus (F) = 0.5 cm enhanced the image quality and assessability. Conclusions: An optimized HRUS examination protocol for quantitative and qualitative facial muscle assessments was proposed.

Background: Genioplasty and chin-augmentation are well-established procedures aimed at enhancing lower facial aesthetics. Traditionally, aesthetic outcomes have been assessed subjectively through expert opinions and patient-reported measures. The integration of artificial intelligence (AI) offers an objective approach to evaluating surgical results. This study utilizes the ICA Aesthetic Navigation AI Research Metrics Model (ICAAN® ARMM) to analyze postoperative changes in facial attractiveness, youthfulness, and skin quality following genioplasty. Methods: Pre- and postoperative full-frontal images of 50 patients undergoing osseous genioplasty were analyzed using the ICAAN® ARMM. Therefore, an array of three aesthetic scores, the Facial Aesthetic Index (FAI), Facial Youthfulness Index (FYI), and Skin Quality Index (SQI), were measured before and after surgery, with subgroup analyses by age, sex, and ethnicity. Minimally clinically important differences (MCIDs) were estimated. Results: All three aesthetic scores demonstrated improvement postoperatively, with FAI showing the greatest increase (82 (73–89) to 85 (75–92); p  = 0.296), without showing statistical significance. Older patients (≥ 35 years) exhibited greater improvements in FAI scores compared to younger individuals (4 (1–10) vs. 1 (− 3–5); p  = 0.028). Sex-related trends were observed, while lacking statistical significance. Ethnic subgroup analysis revealed no differences in score changes, suggesting cross-cultural applicability. Observed improvements did not exceed estimated MCIDs. Conclusion: AI-assisted aesthetic analysis offers a novel, contemporary, and objective method for assessing genioplasty outcomes. While our study suggests general aesthetic improvements following surgery, further research incorporating larger data collections and subjective patient-reported measures is necessary. AI tools hold promise as a complementary tool in aesthetic medicine, supporting both clinicians and patients in surgical decision-making. 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 . Graphical

Background: Androgenetic alopecia, the most common type of hair loss, can impair the patient’s overall mental health. Although there are pharmaceutical and surgical treatments available, little is known about the public standpoint toward hair transplantation (HT). Methods: A sample of individuals living in the United States (n = 1000; male and female participants were equally distributed) was asked to fill out the questionnaire. The online survey was conducted in June 2022. Results: Most participants (42%; n = 416) were between 41 and 60 years of age. Study participants frequently reported that, with their hair loss progressing, they would not feel attractive anymore (n = 400; 40%), nor as confident as before (n = 330; 33%). Although women with minimal hair loss were willing to spend a median price of $4000 [interquartile range (IQ) IQR $1000–$5000], women with extensive hair loss were willing to spend significantly more (median = $5000; IQR $3600–$6375; P = 0.011). This was reproducible in men (P = 0.033). Although significantly fewer women considered undergoing HT (430 women versus 447 men; P < 0.001), female participants were willing to pay more for their HT compared with men (P = 0.039). Conclusions: Individuals living in the United States consider hair loss to impair their attractiveness and regard HT as a valuable therapeutic option. More affordable and gender-specific HT should be subject to future research work.