Section of Dermatology, Department of Health Sciences, University of Genoa, Genoa, Italy.
MaLGa - DIBRIS, University of Genoa, Genoa, Italy.
Br J Dermatol. 2024 Jul 16;191(2):261-266. doi: 10.1093/bjd/ljae142.
BACKGROUND: Artificial intelligence (AI) is reshaping healthcare, using machine and deep learning (DL) to enhance disease management. Dermatology has seen improved diagnostics, particularly in skin cancer detection, through the integration of AI. However, the potential of AI in automating immunofluorescence imaging for autoimmune bullous skin diseases (AIBDs) remains untapped. While direct immunofluorescence (DIF) supports diagnosis, its manual interpretation can hinder efficiency. The use of DL to classify DIF patterns automatically, including the intercellular (ICP) and linear pattern (LP), holds promise for improving the diagnosis of AIBDs. OBJECTIVES: To develop AI algorithms for automated classification of AIBD DIF patterns, such as ICP and LP, in order to enhance diagnostic accuracy, streamline disease management and improve patient outcomes through DL-driven immunofluorescence interpretation. METHODS: We collected immunofluorescence images from skin biopsies of patients suspected of having an AIBD between January 2022 and January 2024. Skin tissue was obtained via a 5-mm punch biopsy, prepared for DIF. Experienced dermatologists classified the images as ICP, LP or negative. To evaluate our DL approach, we divided the images into training (n = 436) and test sets (n = 93). We employed transfer learning with pretrained deep neural networks and conducted fivefold cross-validation to assess model performance. Our dataset's class imbalance was addressed using weighted loss and data augmentation strategies. The models were trained for 50 epochs using Pytorch, achieving an image size of 224 × 224 pixels for both convolutional neural networks (CNNs) and the Swin Transformer. RESULTS: Our study compared six CNNs and the Swin Transformer for AIBD image classification, with the Swin Transformer achieving the highest average validation accuracy (98.5%). On a separate test set, the best model attained an accuracy of 94.6%, demonstrating 95.3% sensitivity and 97.5% specificity across AIBD classes. Visualization with Grad-CAM (class activation mapping) highlighted the model's reliance on characteristic patterns for accurate classification. CONCLUSIONS: The study highlighted the accuracy of CNNs in identifying DIF features. This approach aids automated analysis and reporting, offering reproducibility, speed, data handling and cost-efficiency. Integrating DL into skin immunofluorescence promises precise diagnostics and streamlined reporting in this branch of dermatology.
背景：人工智能（AI）正在重塑医疗保健领域，利用机器学习和深度学习（DL）来增强疾病管理。皮肤科通过整合 AI 提高了诊断水平，特别是在皮肤癌检测方面。然而，AI 自动化免疫荧光成像在自身免疫性大疱性皮肤病（AIBD）中的潜力尚未得到挖掘。直接免疫荧光（DIF）支持诊断，但人工解释可能会降低效率。使用 DL 自动分类 DIF 模式，包括细胞间（ICP）和线性模式（LP），有望提高 AIBD 的诊断准确性。
目的：开发 AI 算法，对 AIBD 的 DIF 模式（如 ICP 和 LP）进行自动分类，通过 DL 驱动的免疫荧光解释提高诊断准确性，简化疾病管理并改善患者预后。
方法：我们收集了 2022 年 1 月至 2024 年 1 月期间疑似患有 AIBD 的患者的皮肤活检免疫荧光图像。通过 5 毫米的打孔活检获得皮肤组织，准备进行 DIF。经验丰富的皮肤科医生将图像分类为 ICP、LP 或阴性。为了评估我们的 DL 方法，我们将图像分为训练集（n = 436）和测试集（n = 93）。我们采用迁移学习和预训练的深度神经网络，并进行五折交叉验证来评估模型性能。我们的数据集中的类别不平衡问题通过加权损失和数据增强策略来解决。使用 Pytorch 对模型进行了 50 个 epoch 的训练，实现了卷积神经网络（CNN）和 Swin Transformer 的图像大小均为 224×224 像素。
结果：我们的研究比较了六个 CNN 和 Swin Transformer 对 AIBD 图像的分类，Swin Transformer 实现了最高的平均验证准确率（98.5%）。在单独的测试集中，最佳模型的准确率达到 94.6%，在 AIBD 类别中具有 95.3%的敏感性和 97.5%的特异性。使用 Grad-CAM（类激活映射）进行可视化突出了模型对准确分类的特征模式的依赖。
结论：该研究强调了 CNN 在识别 DIF 特征方面的准确性。这种方法有助于自动化分析和报告，提供可重复性、速度、数据处理和成本效益。将 DL 整合到皮肤免疫荧光中有望在这一分支的皮肤病学中实现精确诊断和简化报告。
Cochrane Database Syst Rev. 2022-5-20
Health Technol Assess. 2006-9
Cochrane Database Syst Rev. 2022-7-22
J Psychiatr Ment Health Nurs. 2024-8
Cochrane Database Syst Rev. 2022-10-4
Cochrane Database Syst Rev. 2018-1-22
Cochrane Database Syst Rev. 2022-3-2
Health Technol Assess. 2006-6
Zotero 插件