A Deep Learning Approach to Skin Cancer Detection in Dermoscopy Images

This work proposes a deep learning model for skin cancer detection from skin lesion images. In this analytic study, from HAM۱۰۰۰۰ dermoscopy image database, ۳۴۰۰ images were employed including melanoma and non-melanoma lesions. The images comprised ۸۶۰ melanoma, ۳۲۷ actinic keratoses and intraepithelial carcinoma (AKIEC), ۵۱۳ basal cell carcinoma (BCC), ۷۹۵ melanocytic nevi, ۷۹۰ benign keratosis, and ۱۱۵ dermatofibroma cases. A deep convolutional neural network was developed to classify the images into benign and malignant classes. A transfer learning method was leveraged with AlexNet as the pre-trained model. The proposed model takes the raw image as the input and automatically learns useful features from the image for classification. Therefore, it eliminates complex procedures of lesion segmentation and feature extraction. The proposed model achieved an area under the receiver operating characteristic (ROC) curve of ۰.۹۱. Using a confidence score threshold of ۰.۵, a classification accuracy of ۸۴%, the sensitivity of ۸۱%, and specificity of ۸۸% was obtained. The user can change the confidence threshold to adjust sensitivity and specificity if desired. The results indicate the high potential of deep learning for the detection of skin cancer including melanoma and non-melanoma malignancies. The proposed approach can be deployed to assist dermatologists in skin cancer detection. Moreover, it can be applied in smartphones for self-diagnosis of malignant skin lesions. Hence, it may expedite cancer detection that is critical for effective treatment.

کلیدواژه ها:

Skin Cancer ، Deep Learning ، Melanoma ، Transfer Learning ، Dermoscopy