De Logu Francesco, Ugolini Filippo, Maio Vincenza, Simi Sara, Cossu Antonio, Massi Daniela, Nassini Romina, Laurino Marco
Section of Clinical Pharmacology and Oncology, Department of Health Sciences, University of Florence, Florence, Italy.
Section of Pathological Anatomy, Department of Health Sciences, University of Florence, Florence, Italy.
Front Oncol. 2020 Aug 20;10:1559. doi: 10.3389/fonc.2020.01559. eCollection 2020.
Increasing incidence of skin cancer combined with a shortage of dermatopathologists has increased the workload of pathology departments worldwide. In addition, the high intraobserver and interobserver variability in the assessment of melanocytic skin lesions can result in underestimated or overestimated diagnosis of melanoma. Thus, the development of new techniques for skin tumor diagnosis is essential to assist pathologists to standardize diagnoses and plan accurate patient treatment. Here, we describe the development of an artificial intelligence (AI) system that recognizes cutaneous melanoma from histopathological digitalized slides with clinically acceptable accuracy. Whole-slide digital images from 100 formalin-fixed paraffin-embedded primary cutaneous melanoma were used to train a convolutional neural network (CNN) based on a pretrained Inception-ResNet-v2 to accurately and automatically differentiate tumoral areas from healthy tissue. The CNN was trained by using 60 digital slides in which regions of interest (ROIs) of tumoral and healthy tissue were extracted by experienced dermatopathologists, while the other 40 slides were used as test datasets. A total of 1377 patches of healthy tissue and 2141 patches of melanoma were assessed in the training/validation set, while 791 patches of healthy tissue and 1122 patches of pathological tissue were evaluated in the test dataset. Considering the classification by expert dermatopathologists as reference, the trained deep net showed high accuracy (96.5%), sensitivity (95.7%), specificity (97.7%), F score (96.5%), and a Cohen's kappa of 0.929. Our data show that a deep learning system can be trained to recognize melanoma samples, achieving accuracies comparable to experienced dermatopathologists. Such an approach can offer a valuable aid in improving diagnostic efficiency when expert consultation is not available, as well as reducing interobserver variability. Further studies in larger data sets are necessary to verify whether the deep learning algorithm allows subclassification of different melanoma subtypes.
皮肤癌发病率的上升，再加上皮肤病理学家的短缺，增加了全球病理科的工作量。此外，在黑素细胞性皮肤病变评估中，观察者内和观察者间的高变异性可能导致黑色素瘤诊断被低估或高估。因此，开发新的皮肤肿瘤诊断技术对于帮助病理学家规范诊断和规划准确的患者治疗至关重要。在此，我们描述了一种人工智能（AI）系统的开发，该系统能够以临床可接受的准确性从组织病理学数字化切片中识别皮肤黑色素瘤。使用来自100例福尔马林固定石蜡包埋的原发性皮肤黑色素瘤的全切片数字图像，基于预训练的Inception-ResNet-v2训练卷积神经网络（CNN），以准确自动地将肿瘤区域与健康组织区分开来。通过使用60张数字切片对CNN进行训练，其中肿瘤和健康组织的感兴趣区域（ROI）由经验丰富的皮肤病理学家提取，而其他40张切片用作测试数据集。在训练/验证集中共评估了1377个健康组织斑块和2141个黑色素瘤斑块，在测试数据集中评估了791个健康组织斑块和1122个病理组织斑块。以专家皮肤病理学家的分类为参考，训练后的深度网络显示出高准确率（96.5%）、敏感性（95.7%）、特异性（97.7%）、F分数（96.5%）以及Cohen's kappa为0.929。我们的数据表明，可以训练深度学习系统识别黑色素瘤样本，其准确率与经验丰富的皮肤病理学家相当。当无法获得专家咨询时，这种方法可以在提高诊断效率方面提供有价值的帮助，同时减少观察者间的变异性。有必要在更大的数据集中进行进一步研究，以验证深度学习算法是否允许对不同黑色素瘤亚型进行亚分类。
