Unsupervised learning method via triple reconstruction for the classification of ultrasound breast lesions. (August 2022)

• Record Type: Journal Article
• Authors: Song, Mingue; Kim, Yanggon
• Is Part Of: Biomedical signal processing and control. Volume 77(2022)
• Publication Date: 2022-08
• Subjects: Unsupervised learning -- Computer-aid-diagnosis system -- Ultrasound breast tumor classification -- Computer vision; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.103782
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: The proposed method is afforded by a strictly unsupervised fashion in the entire phase. We present an extra training strategy of autoencoder by altering the target mapping of the object. The sequential form of the network enables to encapsulate different details in the embedded space. The experimental results outperformed the supervised learning-based state-of-the-art method. Abstract: Automatic classification of ultrasound breast tumors has been pragmatically alleviating several defects raised in manual diagnosis and utilized as a second opinion to help diagnostic procedures. The extensive works in literature have excessively tackled this task in a supervised learning frame, but the need for large amounts of annotated data followed by such methods represents one of the fatal drawbacks. To address this, this work proposes an unsupervised learning method based on encoder-decoder scheme that allows to avoid the demand of labeling information in the entire discrimination step. A customized convolutional autoencoder is devised as a backbone architecture. The first step of the scenario is to pre-train the network in a generic way. Then, our strategy is to arbitrarily change the target mapping of the object in the reconstruction loss in order to encapsulate different details in the embedding space. Also, the sequential form of autoencoder is introduced to perform gradual reconstruction, and the mutual information derived from three different representations was generated. The discrimination was performed via subsequent clustering, and thus the aim of this study is to suggest the potentiality of unsupervised learning framework in this field. To verify the effectiveness, experiments were conducted on two publicly available datasets, BUSI and UDIAT. The comprehensive results demonstrate that the mutual information is capable of remarkably enhancing discriminant capability and outperforms the actual supervised learning-based state-of-the-art methods in terms of discrimination. Further, our method has advantages of not requiring any region of interest extraction, localization, or annotated information during training.