Del Río-Sancho Sergio, Christen-Zaech Stephanie, Martinez David Alvarez, Pünchera Jöri, Guerrier Stéphane, Laubach Hans J
Laser Dermatology Consultation, Division of Dermatology and Venereology, Geneva University Hospitals, Geneva, Switzerland.
Pediatric Dermatology Unit, Department of Dermatology & Venereology, University Hospital Lausanne, University of Lausanne, Lausanne, Switzerland.
J Eur Acad Dermatol Venereol. 2025 Aug;39(8):1481-1488. doi: 10.1111/jdv.20478. Epub 2024 Dec 17.
Ablative fractional photothermolysis serves as an excellent in vivo model for studying wound healing. The advent of non-invasive imaging devices, such as line-field confocal optical coherence tomography (LC-OCT), enhances this model by enabling detailed monitoring of skin wound healing over time. Additionally, artificial intelligence (AI)-based algorithms are revolutionizing the evaluation of clinical images by providing detailed analyses that are unfeasible manually.
This study aims to assess the value of combining LC-OCT and AI for evaluating the acute wound healing process in the skin.
The forearms of participating volunteers were ablated with a CO laser in a fractional pattern (7.5 mJ/MTZ) (ClinicalTrials.gov identifier: NCT05614557). To induce observable wound healing differences, two different approved silicone-based formulations were randomly assigned to two test sites, with a third site left untreated. In vivo LC-OCT images were obtained at predefined intervals post-laser treatment, ranging from 1 to 7 days. These images were further analysed using AI algorithms.
LC-OCT visualization allows for the characterization of the structural reorganization of the skin during wound healing. The additional integration of AI algorithms significantly enhances the evaluation of the efficacy of wound care interventions by providing a deeper understanding of how these interventions improve wound healing. This is particularly valuable for primary care providers and dermatologists, as AI algorithms have proven useful in observing, characterizing and understanding keratinocyte behaviour.
The combination of AI and high-resolution imaging represents a promising tool for better understanding wound healing, evaluating the efficacy of current wound care interventions and analysing keratinocyte behaviour in detail during the wound healing process.
NCT05614557.
剥脱性点阵光热解是研究伤口愈合的优秀体内模型。线场共聚焦光学相干断层扫描（LC-OCT）等非侵入性成像设备的出现，通过能够随时间详细监测皮肤伤口愈合，增强了该模型。此外，基于人工智能（AI）的算法正在彻底改变临床图像评估，提供手动无法实现的详细分析。
本研究旨在评估结合LC-OCT和AI评估皮肤急性伤口愈合过程的价值。
参与志愿者的前臂用CO激光以点阵模式（7.5 mJ/MTZ）进行消融（ClinicalTrials.gov标识符：NCT05614557）。为了诱导可观察到的伤口愈合差异，将两种不同的已批准的硅基配方随机分配到两个测试部位，第三个部位不进行治疗。在激光治疗后的预定时间间隔（从1到7天）获取体内LC-OCT图像。这些图像使用AI算法进行进一步分析。
LC-OCT可视化能够表征伤口愈合过程中皮肤的结构重组。AI算法的额外整合通过更深入地了解这些干预措施如何改善伤口愈合，显著增强了对伤口护理干预效果的评估。这对初级保健提供者和皮肤科医生特别有价值，因为AI算法已被证明在观察、表征和理解角质形成细胞行为方面很有用。
AI与高分辨率成像的结合是一种有前途的工具，可用于更好地理解伤口愈合、评估当前伤口护理干预的效果以及在伤口愈合过程中详细分析角质形成细胞行为。
NCT05614557。
Zotero 插件