Ren Togo1, Haruna Watanabe2, Takahiro Ogawa3, Miki Haseyama4. 1. Education and Research Center for Mathematical and Data Science, Hokkaido University, N-12, W-7, Kita-ku, Sapporo, 060-0812, Japan. Electronic address: togo@lmd.ist.hokudai.ac.jp. 2. Faculty of Information Science and Technology, Hokkaido University, N-14, W-9, Kita-Ku, Sapporo, 060-0814, Japan. Electronic address: haruna@lmd.ist.hokudai.ac.jp. 3. Faculty of Information Science and Technology, Hokkaido University, N-14, W-9, Kita-Ku, Sapporo, 060-0814, Japan. Electronic address: ogawa@lmd.ist.hokudai.ac.jp. 4. Faculty of Information Science and Technology, Hokkaido University, N-14, W-9, Kita-Ku, Sapporo, 060-0814, Japan. Electronic address: miki@ist.hokudai.ac.jp.
Abstract
AIM: The aim of this study was to determine whether our deep convolutional neural network-based anomaly detection model can distinguish differences in esophagus images and stomach images obtained from gastric X-ray examinations. METHODS: A total of 6012 subjects were analyzed as our study subjects. Since the number of esophagus X-ray images is much smaller than the number of gastric X-ray images taken in X-ray examinations, we took an anomaly detection approach to realize the task of organ classification. We constructed a deep autoencoding gaussian mixture model (DAGMM) with a convolutional autoencoder architecture. The trained model can produce an anomaly score for a given test X-ray image. For comparison, the original DAGMM, AnoGAN, and a One-Class Support Vector Machine (OCSVM) that were trained with features obtained by a pre-trained Inception-v3 network were used. RESULTS: Sensitivity, specificity, and the calculated harmonic mean of the proposed method were 0.956, 0.980, and 0.968, respectively. Those of the original DAGMM were 0.932, 0.883, and 0.907, respectively. Those of AnoGAN were 0.835, 0.833, and 0.834, respectively, and those of OCSVM were 0.932, 0.935, and 0.934, respectively. Experimental results showed the effectiveness of the proposed method for an organ classification task. CONCLUSION: Our deep convolutional neural network-based anomaly detection model has shown the potential for clinical use in organ classification.
AIM: The aim of this study was to determine whether our deep convolutional neural network-based anomaly detection model can distinguish differences in esophagus images and stomach images obtained from gastric X-ray examinations. METHODS: A total of 6012 subjects were analyzed as our study subjects. Since the number of esophagus X-ray images is much smaller than the number of gastric X-ray images taken in X-ray examinations, we took an anomaly detection approach to realize the task of organ classification. We constructed a deep autoencoding gaussian mixture model (DAGMM) with a convolutional autoencoder architecture. The trained model can produce an anomaly score for a given test X-ray image. For comparison, the original DAGMM, AnoGAN, and a One-Class Support Vector Machine (OCSVM) that were trained with features obtained by a pre-trained Inception-v3 network were used. RESULTS: Sensitivity, specificity, and the calculated harmonic mean of the proposed method were 0.956, 0.980, and 0.968, respectively. Those of the original DAGMM were 0.932, 0.883, and 0.907, respectively. Those of AnoGAN were 0.835, 0.833, and 0.834, respectively, and those of OCSVM were 0.932, 0.935, and 0.934, respectively. Experimental results showed the effectiveness of the proposed method for an organ classification task. CONCLUSION: Our deep convolutional neural network-based anomaly detection model has shown the potential for clinical use in organ classification.