Xiaoyu Cui1, Ran Wei1, Lixin Gong1, Ruiqun Qi2, Zeyin Zhao1, Hongduo Chen3, Kaixin Song1, Amer A A Abdulrahman3, Yining Wang3, John Z S Chen4, Shuo Chen1, Yue Zhao1, Xinghua Gao3. 1. Department of Biomedical Informatics, College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China. 2. NHC Key Laboratory of Immunodermatology (China Medical University), Ministry of Education Key Laboratory of Immunodermatology (China Medical University), Department of Dermatology The First Hospital of China Medical University, Shenyang, China. Electronic address: xiaoqiliumin@163.com. 3. NHC Key Laboratory of Immunodermatology (China Medical University), Ministry of Education Key Laboratory of Immunodermatology (China Medical University), Department of Dermatology The First Hospital of China Medical University, Shenyang, China. 4. HealthySkin Dermatology, LLP, Tucson, Arizona.
Abstract
BACKGROUND: Artificial intelligence methods for the classification of melanoma have been studied extensively. However, few studies compare these methods under the same standards. OBJECTIVE: To seek the best artificial intelligence method for diagnosis of melanoma. METHODS: The contrast test used 2200 dermoscopic images. Image segmentations, feature extractions, and classifications were performed in sequence for evaluation of traditional machine learning algorithms. The recent popular convolutional neural network frameworks were used for transfer learning training classification. RESULTS: The region growing algorithm has the best segmentation performance, with an intersection over union of 70.06% and a false-positive rate of 17.67%. Classification performance was better with logistic regression, with a sensitivity of 76.36% and a specificity of 87.04%. The Inception V3 model (Google, Mountain View, CA) worked best in deep learning algorithms: the accuracy was 93.74%, the sensitivity was 94.36%, and the specificity was 85.64%. LIMITATIONS: There was no division in the severity of melanoma samples used in this experiment. The data set was relatively small for deep learning. CONCLUSION: The performance of traditional machine learning is satisfactory for the small data set of melanoma dermoscopic images, and the potential for deep learning in the future big data era is enormous.
BACKGROUND: Artificial intelligence methods for the classification of melanoma have been studied extensively. However, few studies compare these methods under the same standards. OBJECTIVE: To seek the best artificial intelligence method for diagnosis of melanoma. METHODS: The contrast test used 2200 dermoscopic images. Image segmentations, feature extractions, and classifications were performed in sequence for evaluation of traditional machine learning algorithms. The recent popular convolutional neural network frameworks were used for transfer learning training classification. RESULTS: The region growing algorithm has the best segmentation performance, with an intersection over union of 70.06% and a false-positive rate of 17.67%. Classification performance was better with logistic regression, with a sensitivity of 76.36% and a specificity of 87.04%. The Inception V3 model (Google, Mountain View, CA) worked best in deep learning algorithms: the accuracy was 93.74%, the sensitivity was 94.36%, and the specificity was 85.64%. LIMITATIONS: There was no division in the severity of melanoma samples used in this experiment. The data set was relatively small for deep learning. CONCLUSION: The performance of traditional machine learning is satisfactory for the small data set of melanoma dermoscopic images, and the potential for deep learning in the future big data era is enormous.