深度学习算法在新鲜切除组织的离体共聚焦显微镜图像中分类基底细胞癌
本研究利用新鲜切除组织的离体共聚焦显微镜(EVCM)图像,训练深度学习算法自动检测基底细胞癌(BCC)。研究纳入40例BCC和28例阴性样本,构建EVCM、H&E及联合数据集。EVCM训练模型诊断准确率达92%,与H&E模型(93%)相似,ROC曲线下面积分别为0.94、0.95和0.94。该算法有望辅助Mohs手术中的BCC检测。
要点
- EVCM深度学习模型检测BCC准确率92%,与皮肤科医生相当。
- 仅用H&E图像训练的模型也能准确检测EVCM图像中的BCC。
- 该算法可辅助Mohs手术中快速检测BCC。
核心结论
- 深度学习算法在EVCM图像中自动检测BCC的准确率与皮肤科医生相当,可辅助Mohs手术中的诊断。
研究设计与样本
- 研究收集40例BCC和28例阴性样本,构建EVCM图像数据集(663张)、H&E图像数据集(516张)及联合数据集用于训练。
- 另收集7例BCC和4例阴性样本构建EVCM测试数据集(107张图像)。
主要发现
- EVCM训练模型诊断准确率为92%,H&E模型为93%。
- EVCM、H&E和联合训练模型的ROC曲线下面积分别为0.94、0.95和0.94。
- 仅用H&E图像训练的模型也能准确检测EVCM图像中的BCC。
临床怎么用
- 该算法可用于Mohs手术中辅助快速检测BCC,减少对专业培训的依赖。
- H&E图像资源更丰富,可用于训练模型以检测EVCM图像中的BCC。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
Ex vivo confocal microscopy (EVCM) generates digitally colored purple-pink images similar to H&E without time-consuming tissue processing. It can be used during Mohs surgery for rapid detection of basal cell carcinoma (BCC); however, reading EVCM images requires specialized training. An automated approach using a deep learning algorithm for BCC detection in EVCM images can aid in diagnosis. A total of 40 BCCs and 28 negative (not-BCC) samples were collected at Memorial Sloan Kettering Cancer Center to create three training datasets: (i) EVCM image dataset (663 images), (ii) H&E image dataset (516 images), and (iii) a combination of the two datasets. A total of seven BCCs and four negative samples were collected to create an EVCM test dataset (107 images). The model trained with the EVCM dataset achieved 92% diagnostic accuracy, similar to the H&E model (93%). The area under the receiver operator characteristic curve was 0.94, 0.95, and 0.94 for EVCM-, H&E-, and combination-trained models, respectively. We developed an algorithm for automatic BCC detection in EVCM images (comparable accuracy to dermatologists). This approach could be used to assist with BCC detection during Mohs surgery. Furthermore, we found that a model trained with only H&E images (which are more available than EVCM images) can accurately detect BCC in EVCM images.
引用
- Sendín-Martín M, Lara-Caro M, Harris U, et al. Classification of Basal Cell Carcinoma in Ex Vivo Confocal Microscopy Images from Freshly Excised Tissues Using a Deep Learning Algorithm. J Invest Dermatol. 2021. PMID: 34695413. doi:10.1016/j.jid.2021.09.029
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。