日本語フィールド
著者：Ryuji Ohura, Hajime Omura. ,Yasuhisa Sakata, Teruya MInamoto題名：Computer-aided diagnosis method for detecting early esophageal cancer from endoscopic image by using dyadic wavelet transform and fractal dimension発表情報：Advances in Intelligent Systems and Computing 巻: 448 ページ: 929-938キーワード：概要：抄録：We propose a new computer-aided method for diagnosing early esophageal cancer from endoscopic images by using the dyadic wavelet transform (DYWT) and the fractal dimension. In our method, an input image is converted into HSV color space, and a fusion image is made from the S (saturation) and V (value) components based on the DYWT. We apply the contrast enhancement to produce a grayscale image in which the structure of abnormal regions is enhanced. We can obtain binary images composed of multiple layers by low-gradation processing. We visualize abnormal regions by summing these fractal dimensions by computing the complexity of these images. We describe a process for enhancing, detecting and visualizing abnormal regions in detail, and we present experimental results demonstrating that our method gives visualized images in which abnormal regions in endoscopic images can be located and that contain data useful for actual diagnosis of early esophageal cancer.英語フィールド
Author：Ryuji Ohura, Hajime Omura. ,Yasuhisa Sakata, Teruya MInamotoTitle：Computer-aided diagnosis method for detecting early esophageal cancer from endoscopic image by using dyadic wavelet transform and fractal dimensionAnnouncement information:Advances in Intelligent Systems and Computing Vol: 448 Page: 929-938An abstract：We propose a new computer-aided method for diagnosing early esophageal cancer from endoscopic images by using the dyadic wavelet transform (DYWT) and the fractal dimension. In our method, an input image is converted into HSV color space, and a fusion image is made from the S (saturation) and V (value) components based on the DYWT. We apply the contrast enhancement to produce a grayscale image in which the structure of abnormal regions is enhanced. We can obtain binary images composed of multiple layers by low-gradation processing. We visualize abnormal regions by summing these fractal dimensions by computing the complexity of these images. We describe a process for enhancing, detecting and visualizing abnormal regions in detail, and we present experimental results demonstrating that our method gives visualized images in which abnormal regions in endoscopic images can be located and that contain data useful for actual diagnosis of early esophageal cancer.