Artificial Intelligence Deep Learning Ultrasound Discrimination of Cosmetic Fillers: A Multicenter Study.
Authors
Affiliations (16)
Affiliations (16)
- Department of Dermatology, Faculty of Medicine, Universidad de Chile, Santiago, Chile.
- Department of Dermatology, School of Medicine, Pontificia Universidad Catolica de Chile, Santiago, Chile.
- Department of Dermatology and Cutaneous Surgery, Miller School of Medicine, University of Miami, Miami, Florida, USA.
- Institute for Diagnostic Imaging and Research of the Skin and Soft Tissues, Santiago, Chile.
- Department of Computer Science and Artificial Intelligence, Universidad de Granada, Granada, Spain.
- Department of Radiology, Hospital Sirio Libanes, São Paulo, Brazil.
- Radioderm SP, São Paulo, Brazil.
- Centre Mèdic de Cabo Bové, Martorell, Spain.
- ED Ultrassonografia Dermatologica, São Paulo, Brazil.
- Department of Ultrasound, Institute of Radiology, Hospital das Clínicas, Medical School of University of São Paulo, São Paulo, Brazil.
- Ultrassonando Clinic, São Paulo, Brazil.
- Ultraplenna Clinic, Recife, Brazil.
- Department of Dermatology, Hospital of Gamboa, IMS, Rio de Janeiro, Brazil.
- Highly Specialized Ultrasound Center, Bogotá, Colombia.
- Department of Dermatology, Erasmus University Medical Center, Rotterdam, The Netherlands.
- Aesthetic Clinic, Entre Ríos, Argentina.
Abstract
Despite the growing use of artificial intelligence (AI) in medicine, imaging, and dermatology, to date, there is no information on the use of AI for discriminating cosmetic fillers on ultrasound (US). An international collaborative group working in dermatologic and esthetic US was formed and worked with the staff of the Department of Computer Science and AI of the Universidad de Granada to gather and process a relevant number of anonymized images. AI techniques based on deep learning (DL) with YOLO (you only look once) architecture, together with a bounding box annotation tool, allowed experts to manually delineate regions of interest for the discrimination of common cosmetic fillers under real-world conditions. A total of 14 physicians from 6 countries participated in the AI study and compiled a final dataset comprising 1432 US images, including HA (hyaluronic acid), PMMA (polymethylmethacrylate), CaHA (calcium hydroxyapatite), and SO (silicone oil) filler cases. The model exhibits robust and consistent classification performance, with an average accuracy of 0.92 ± 0.04 across the cross-validation folds. YOLOv11 demonstrated outstanding performance in the detection of HA and SO, yielding F1 scores of 0.96 ± 0.02 and 0.94 ± 0.04, respectively. On the other hand, CaHA and PMMA show somewhat lower and less consistent performance in terms of precision and recall, with F1-scores around 0.83. AI using YOLOv11 allowed us to discriminate reliably between HA and SO using different complexity high-frequency US devices and operators. Further AI DL-specific work is needed to identify CaHA and PMMA more accurately.