Abstract
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%), F1 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.
Highlights
Melanoma is one of the major causes of cancer-related death, and its incidence is increasing worldwide [1, 2]
Our data show that a deep learning system can be trained to recognize melanoma samples, achieving accuracies comparable to experienced dermatopathologists
We showed that an artificial intelligence (AI) algorithm to recognize cutaneous melanoma can be a useful diagnostic tool for supporting dermatopathologists for diagnostic purposes
Summary
Melanoma is one of the major causes of cancer-related death, and its incidence is increasing worldwide [1, 2]. A lack of access to dermatopathology expertise in this context can slow diagnostic turnaround times, resulting in delays in patient care and leading to potential adverse impacts on clinical outcomes. In this scenario, the computer-aided diagnosis (CAD) system reduces intraobserver and interobserver variability and improves the accuracy of pathology interpretation [5, 6]. The training of a new CNN model with a new image dataset requires extensive effort to collect a large number of images For this reason, a CNN architecture can be built from pretrained models (Transfer Learning approach) with a considerable reduction in the training image dataset [11]. Several types of pretrained CNN architectures, including AlexNet, SqueezeNet, NASNet-Large, Inception-v3, ResNet-50, Vgg, and Inception- ResNet-v2 [12,13,14], have been designed
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
