{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T11:15:21Z","timestamp":1783163721770,"version":"3.54.6"},"reference-count":43,"publisher":"Oxford University Press (OUP)","license":[{"start":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T00:00:00Z","timestamp":1781308800000},"content-version":"vor","delay-in-days":163,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,1,14]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Multimodal large language models (MLLMs) are increasingly applied to visual assessment tasks, yet their ability to evaluate aesthetic treatment outcomes remains unclear.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>The aim of this study was to assess whether contemporary MLLMs can identify treatment state and detect region-specific aesthetic improvement following botulinum neurotoxin treatment.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>In this observational study, 23 paired facial image cases (46 images; 460 model evaluations) were analyzed. Four MLLMs (GPT-5.4 Pro [OpenAI, San Francisco, CA], Grok 4.1 [xAI, San Francisco, CA], Gemini 3.1 Pro [Google DeepMind, Mountain View, CA], and Claude Opus 4.6 [Anthropic, San Francisco, CA]) performed 5 independent inference runs per case. Models identified the posttreatment image and assessed regional improvement (forehead, glabella, and periorbital). Accuracy, sensitivity, specificity, balanced accuracy, Matthews correlation coefficient, and Fleiss\u2019 \u03ba were calculated descriptively. Performance was compared with majority-class baselines. Exploratory outputs included aesthetic scores and apparent age estimates.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>All models identified the posttreatment image (100% accuracy). Region-specific improvement detection frequently failed to exceed majority-class baselines (65.2%-91.3%). Gemini 3.1 Pro showed the highest performance for the forehead (74.8%) and glabella (63.5%), whereas no model reached the periorbital baseline. Inter-run reliability varied widely (\u03ba \u22120.113 to 0.719). High reliability did not imply correctness. All models systematically overestimated improvement (62.6%-94.8% of predictions exceeded ground truth). False positives exceeded false negatives in all 12 model\u2013task combinations. Exploratory outputs indicated perceived rejuvenation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>MLLMs recognize the format of aesthetic change but not its clinical nuance. Bridging this gap requires more than improved accuracy: a coordinated agenda of domain-specific fine-tuning, expert-rater benchmarking, and structured outcome frameworks, supported by governance of training-data provenance and clear safeguards before any clinical or research deployment.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Level of Evidence: 5 (Therapeutic)<\/jats:title>\n                    <jats:p>For image description, please refer to the figure legend and surrounding text.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/asjof\/ojag112","type":"journal-article","created":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T11:47:04Z","timestamp":1781092024000},"source":"Crossref","is-referenced-by-count":0,"title":["Evaluation of Aesthetic Outcomes Following Botulinum Toxin Treatment Using Multimodal Large Language Models: A Paired Before-and-After Analysis"],"prefix":"10.1093","volume":"8","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6395-6670","authenticated-orcid":false,"given":"Ibrahim","family":"G\u00fcler","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Armin","family":"Kraus","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gerrit","family":"Grieb","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Henrik","family":"Stelling","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,6,13]]},"reference":[{"key":"2026070406164248100_ojag112-B1","doi-asserted-by":"publisher","first-page":"182","DOI":"10.1016\/j.clindermatol.2007.09.008","article-title":"Advanced botulinum toxin techniques against wrinkles in the upper face","volume":"26","author":"Salti","year":"2008","journal-title":"Clin Dermatol"},{"key":"2026070406164248100_ojag112-B2","doi-asserted-by":"publisher","DOI":"10.1097\/GOX.0000000000004228","article-title":"The multispecialty toxin: a literature review of botulinum toxin","volume":"10","author":"Bach","year":"2022","journal-title":"Plast Reconstr Surg Glob Open"},{"key":"2026070406164248100_ojag112-B3","doi-asserted-by":"publisher","first-page":"265e","DOI":"10.1097\/PRS.0000000000003544","article-title":"Aesthetic leaders in facial aesthetics consensus committee. Facial assessment and injection guide for botulinum toxin and injectable hyaluronic acid fillers: focus on the upper face","volume":"140","author":"de Maio","year":"2017","journal-title":"Plast Reconstr Surg"},{"key":"2026070406164248100_ojag112-B4","doi-asserted-by":"publisher","first-page":"350","DOI":"10.1097\/SAP.0b013e31824a43e0","article-title":"Subjective rating of cosmetic treatment with botulinum toxin type A: do existing measures demonstrate interobserver validity?","volume":"69","author":"Conkling","year":"2012","journal-title":"Ann Plast Surg"},{"key":"2026070406164248100_ojag112-B5","doi-asserted-by":"publisher","first-page":"303e","DOI":"10.1097\/PRS.0000000000008708","article-title":"Tips and tricks for facial toxin injections with illustrated anatomy","volume":"149","author":"Swift","year":"2022","journal-title":"Plast Reconstr Surg"},{"key":"2026070406164248100_ojag112-B6","doi-asserted-by":"publisher","first-page":":","DOI":"10.1093\/asjof\/ojaf032","article-title":"Botulinum toxin: surely, we can do better? Optimizing results beyond on-label techniques and teaching","volume":"7","author":"Goodman","year":"2025","journal-title":"Aesthet Surg J Open Forum"},{"key":"2026070406164248100_ojag112-B7","doi-asserted-by":"publisher","author":"Vaswani","year":"2017","DOI":"10.48550\/arXiv.1706.03762"},{"key":"2026070406164248100_ojag112-B8","doi-asserted-by":"publisher","DOI":"10.1093\/nsr\/nwae403","article-title":"A survey on multimodal large language models","volume":"11","author":"Yin","year":"2024","journal-title":"Natl Sci Rev"},{"key":"2026070406164248100_ojag112-B9","doi-asserted-by":"publisher","first-page":"900","DOI":"10.3348\/kjr.2025.0599","article-title":"Multimodal large language models in medical imaging: current state and future directions","volume":"26","author":"Nam","year":"2025","journal-title":"Korean J Radiol"},{"key":"2026070406164248100_ojag112-B10","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.250617","article-title":"One year on: assessing progress of multimodal large language model performance on RSNA 2024 case of the day questions","volume":"316","author":"Hou","year":"2025","journal-title":"Radiology"},{"key":"2026070406164248100_ojag112-B11","doi-asserted-by":"publisher","DOI":"10.1097\/MD.0000000000043716","article-title":"Diagnostic performance of ChatGPT-4.0 in elbow fracture detection: a comparative study of radial head, distal humerus, and olecranon fractures","volume":"104","author":"G\u00fcltekin","year":"2025","journal-title":"Medicine (Baltimore)"},{"key":"2026070406164248100_ojag112-B12","doi-asserted-by":"publisher","first-page":"360","DOI":"10.1097\/BOT.0000000000003166","article-title":"Comparing the efficacy between ChatGPT 5, Grok 3, and Claude 4.5 sonnet in analyzing orthopedic trauma-related imaging","volume":"40","author":"Holmstrom","year":"2026","journal-title":"J Orthop Trauma"},{"key":"2026070406164248100_ojag112-B13","doi-asserted-by":"publisher","first-page":"311","DOI":"10.1111\/ddg.15609","article-title":"The use of a ChatGPT-4-based chatbot in teledermatology: a retrospective exploratory study","volume":"23","author":"Shapiro","year":"2025","journal-title":"J Dtsch Dermatol Ges"},{"key":"2026070406164248100_ojag112-B14","doi-asserted-by":"publisher","DOI":"10.1007\/s00266-026-05676-8","article-title":"ChatGPT-5 matches surgeon-level assessment of facelift candidacy: a pilot proof-of-concept study","author":"Saeed","year":"2026","journal-title":"Aesthetic Plast Surg"},{"key":"2026070406164248100_ojag112-B15","doi-asserted-by":"publisher","first-page":"NP1","DOI":"10.1093\/asj\/sjae204","article-title":"Evaluating the impact of BoNT-A injections on facial expressions: a deep learning analysis","volume":"45","author":"Aktar Ugurlu","year":"2024","journal-title":"Aesthet Surg J"},{"key":"2026070406164248100_ojag112-B16","doi-asserted-by":"publisher","DOI":"10.1093\/asj\/sjag037","article-title":"Botulinum toxin and social perception: effects of upper-face treatment on facial emotion attribution","author":"Eilert","year":"2026","journal-title":"Aesthet Surg J"},{"key":"2026070406164248100_ojag112-B17","doi-asserted-by":"publisher","DOI":"10.3390\/jcm14103484","article-title":"Facial analysis for plastic surgery in the era of artificial intelligence: a comparative evaluation of multimodal large language models","volume":"14","author":"Haider","year":"2025","journal-title":"J Clin Med"},{"key":"2026070406164248100_ojag112-B18","doi-asserted-by":"publisher","author":"Anggriani","DOI":"10.64898\/2025.12.08.25341823"},{"key":"2026070406164248100_ojag112-B19","doi-asserted-by":"publisher","author":"Shyr","DOI":"10.1101\/2025.08.06.25333170"},{"key":"2026070406164248100_ojag112-B20","doi-asserted-by":"publisher","author":"Alvarado Gonzalez","year":"2025","DOI":"10.48550\/arXiv.2509.24086"},{"key":"2026070406164248100_ojag112-B21","doi-asserted-by":"publisher","first-page":"378","DOI":"10.1037\/h0031619","article-title":"Measuring nominal scale agreement among many raters","volume":"76","author":"Fleiss","year":"1971","journal-title":"Psychol Bull"},{"key":"2026070406164248100_ojag112-B22"},{"key":"2026070406164248100_ojag112-B23","doi-asserted-by":"publisher","first-page":"159","DOI":"10.2307\/2529310","article-title":"The measurement of observer agreement for categorical data","volume":"33","author":"Landis","year":"1977","journal-title":"Biometrics"},{"key":"2026070406164248100_ojag112-B24","doi-asserted-by":"publisher","first-page":"665","DOI":"10.1038\/s42256-020-00257-z","article-title":"Shortcut learning in deep neural networks","volume":"2","author":"Geirhos","year":"2020","journal-title":"Nat Mach Intell"},{"key":"2026070406164248100_ojag112-B25","doi-asserted-by":"publisher","first-page":"1395","DOI":"10.1093\/asj\/sjad135","article-title":"Unleashing the power of ChatGPT: revolutionizing plastic surgery and beyond","volume":"43","author":"Bassiri-Tehrani","year":"2023","journal-title":"Aesthet Surg J"},{"key":"2026070406164248100_ojag112-B26","doi-asserted-by":"publisher","first-page":"113","DOI":"10.1093\/asj\/sjaf239","article-title":"Artificial intelligence in plastic surgery: current status, limitations, and future directions","volume":"46","author":"Copeland-Halperin","year":"2026","journal-title":"Aesthet Surg J"},{"key":"2026070406164248100_ojag112-B27","doi-asserted-by":"publisher","first-page":"45","DOI":"10.1097\/PRS.0000000000008020","article-title":"Turning back the clock: artificial intelligence recognition of age reduction after face-lift surgery correlates with patient satisfaction","volume":"148","author":"Zhang","year":"2021","journal-title":"Plast Reconstr Surg"},{"key":"2026070406164248100_ojag112-B28","doi-asserted-by":"publisher","first-page":"2407","DOI":"10.1007\/s00266-023-03538-1","article-title":"Complications following facelift and neck lift: implementation and assessment of large language model and artificial intelligence (ChatGPT) performance across 16 simulated patient presentations","volume":"47","author":"Abi-Rafeh","year":"2023","journal-title":"Aesthetic Plast Surg"},{"key":"2026070406164248100_ojag112-B29","doi-asserted-by":"publisher","DOI":"10.1093\/asjof\/ojae058","article-title":"Comparative performance of current patient-accessible artificial intelligence large language models in the preoperative education of patients in facial aesthetic surgery","volume":"6","author":"Abi-Rafeh","year":"2024","journal-title":"Aesthet Surg J Open Forum"},{"key":"2026070406164248100_ojag112-B30","doi-asserted-by":"publisher","first-page":"889","DOI":"10.1093\/asj\/sjae025","article-title":"Utility and comparative performance of current artificial intelligence large language models as postoperative medical support chatbots in aesthetic surgery","volume":"44","author":"Abi-Rafeh","year":"2024","journal-title":"Aesthet Surg J"},{"key":"2026070406164248100_ojag112-B31","doi-asserted-by":"publisher","first-page":"115","DOI":"10.3390\/toxins18020115","article-title":"From muscular hypertonus to equilibrium: a conceptual framework for aesthetic neuromodulation based on the Index of Muscular Equilibrium (IME)","volume":"18","author":"Armenti","year":"2026","journal-title":"Toxins (Basel)"},{"key":"2026070406164248100_ojag112-B32","author":"Bundesgerichtshof"},{"key":"2026070406164248100_ojag112-B33","doi-asserted-by":"publisher","DOI":"10.1016\/j.patter.2025.101260","article-title":"Unleashing the potential of prompt engineering for large language models","volume":"6","author":"Chen","year":"2025","journal-title":"Patterns (N Y)"},{"key":"2026070406164248100_ojag112-B34","doi-asserted-by":"publisher","first-page":"37932","DOI":"10.1038\/s41598-025-22979-z","article-title":"Prompt-dependent performance of multimodal AI model in oral diagnosis: a comprehensive analysis of accuracy, narrative quality, calibration, and latency versus human experts","volume":"15","author":"Hassanein","year":"2025","journal-title":"Sci Rep"},{"key":"2026070406164248100_ojag112-B35","doi-asserted-by":"publisher","DOI":"10.3389\/frai.2025.1658316","article-title":"Prompt engineering for accurate statistical reasoning with large language models in medical research","volume":"8","author":"Vilakati","year":"2025","journal-title":"Front Artif Intell"},{"key":"2026070406164248100_ojag112-B36","doi-asserted-by":"publisher","first-page":"223","DOI":"10.1093\/asj\/sjae219","article-title":"Artificial intelligence, data protection, privacy, and doxxing","volume":"45","author":"Copeland-Halperin","year":"2025","journal-title":"Aesthet Surg J"},{"key":"2026070406164248100_ojag112-B37","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1097\/00006534-200111000-00048","article-title":"A classification of facial wrinkles","volume":"108","author":"Lemperle","year":"2001","journal-title":"Plast Reconstr Surg"},{"key":"2026070406164248100_ojag112-B38","doi-asserted-by":"publisher","first-page":"134","DOI":"10.1016\/s1085-5629(96)80003-4","article-title":"Aesthetic and anatomic analysis of the aging skin","volume":"15","author":"Glogau","year":"1996","journal-title":"Semin Cutan Med Surg"},{"key":"2026070406164248100_ojag112-B39","doi-asserted-by":"publisher","first-page":"1249","DOI":"10.1111\/bjd.13509","article-title":"Evaluation of skin ageing: a systematic review of clinical scales","volume":"172","author":"Dobos","year":"2015","journal-title":"Br J Dermatol"},{"key":"2026070406164248100_ojag112-B40","doi-asserted-by":"publisher","first-page":"1175","DOI":"10.1093\/asj\/sjac108","article-title":"A correlation of the glogau scale with VISIA-CR complexion analysis measurements in assessing facial photoaging for clinical research","volume":"42","author":"Oesch","year":"2022","journal-title":"Aesthet Surg J"},{"key":"2026070406164248100_ojag112-B41","doi-asserted-by":"publisher","DOI":"10.1093\/asjof\/ojab059","article-title":"A single-center pilot study to classify signs of dorsal hand aging using 3 grading scales","volume":"4","author":"Wamsley","year":"2022","journal-title":"Aesthet Surg J Open Forum"},{"key":"2026070406164248100_ojag112-B42","doi-asserted-by":"publisher","first-page":"235","DOI":"10.3109\/14764172.2010.514920","article-title":"A validated facial grading scale: the future of facial ageing measurement tools?","volume":"12","author":"Carruthers","year":"2010","journal-title":"J Cosmet Laser Ther"},{"key":"2026070406164248100_ojag112-B43","doi-asserted-by":"publisher","first-page":"5978","DOI":"10.5826\/dpc.1601a5978","article-title":"Artificial intelligence versus human assessment in the treatment of upper facial wrinkles with abobotulinum toxin A","volume":"16","author":"K\u00fc\u00e7\u00fck","year":"2026","journal-title":"Dermatol Pract Concept"}],"container-title":["Aesthetic Surgery Journal Open Forum"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/asjopenforum\/advance-article-pdf\/doi\/10.1093\/asjof\/ojag112\/68520947\/ojag112.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/asjopenforum\/article-pdf\/doi\/10.1093\/asjof\/ojag112\/68520947\/ojag112.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/asjopenforum\/article-pdf\/doi\/10.1093\/asjof\/ojag112\/68520947\/ojag112.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T10:16:49Z","timestamp":1783160209000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/asjopenforum\/article\/doi\/10.1093\/asjof\/ojag112\/8707414"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":43,"URL":"https:\/\/doi.org\/10.1093\/asjof\/ojag112","relation":{},"ISSN":["2631-4797"],"issn-type":[{"value":"2631-4797","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026]]},"published":{"date-parts":[[2026]]},"article-number":"ojag112"}}