Xinle Wang1, Haiyang Qian1, Edward J Ciaccio2, Suzanne K Lewis2, Govind Bhagat3, Peter H Green2, Shenghao Xu4, Liang Huang1, Rongke Gao5, Yu Liu6. 1. School of Instrument Science and Opto-electronic Engineering, Hefei University of Technology, Hefei 230009, China. 2. Columbia University Medical Center, Department of Medicine - Celiac Disease Center, New York, USA. 3. Columbia University Medical Center, Department of Medicine - Celiac Disease Center, New York, USA; Columbia University Medical Center, Department of Pathology and Cell Biology, New York, USA. 4. Shandong Key Laboratory of Biochemical Analysis, College of Chemistry and Molecular Engineering, Qingdao University of Science and Technology, Qingdao 266042, China. 5. School of Instrument Science and Opto-electronic Engineering, Hefei University of Technology, Hefei 230009, China. Electronic address: rkgao@hfut.edu.cn. 6. School of Instrument Science and Opto-electronic Engineering, Hefei University of Technology, Hefei 230009, China. Electronic address: yuliu@hfut.edu.cn.
Abstract
BACKGROUND AND OBJECTIVE: Videocapsule endoscopy (VCE) is a relatively new technique for evaluating the presence of villous atrophy in celiac disease patients. The diagnostic analysis of video frames is currently time-consuming and tedious. Recently, computer-aided diagnosis (CAD) systems have become an attractive research area for diagnosing celiac disease. However, the images captured from VCE are susceptible to alterations in light illumination, rotation direction, and intestinal secretions. Moreover, textural features of the mucosal villi obtained by VCE are difficult to characterize and extract. This work aims to find a novel deep learning feature learning module to assist in the diagnosis of celiac disease. METHODS: In this manuscript, we propose a novel deep learning recalibration module which shows significant gain in diagnosing celiac disease. In this recalibration module, the block-wise recalibration component is newly employed to capture the most salient feature in the local channel feature map. This learning module was embedded into ResNet50, Inception-v3 to diagnose celiac disease using a 10-time 10-fold cross-validation based upon analysis of VCE images. In addition, we employed model weights to extract feature points from training and test samples before the last fully connected layer, and then input to a support vector machine (SVM), k-nearest neighbor (KNN), and linear discriminant analysis (LDA) for differentiating celiac disease images from heathy controls. RESULTS: Overall, the accuracy, sensitivity and specificity of the 10-time 10-fold cross-validation were 95.94%, 97.20% and 95.63%, respectively. CONCLUSIONS: A novel deep learning recalibration module, with global response and local salient factors is proposed, and it has a high potential for utilizing deep learning networks to diagnose celiac disease using VCE images.
BACKGROUND AND OBJECTIVE: Videocapsule endoscopy (VCE) is a relatively new technique for evaluating the presence of villous atrophy in celiac diseasepatients. The diagnostic analysis of video frames is currently time-consuming and tedious. Recently, computer-aided diagnosis (CAD) systems have become an attractive research area for diagnosing celiac disease. However, the images captured from VCE are susceptible to alterations in light illumination, rotation direction, and intestinal secretions. Moreover, textural features of the mucosal villi obtained by VCE are difficult to characterize and extract. This work aims to find a novel deep learning feature learning module to assist in the diagnosis of celiac disease. METHODS: In this manuscript, we propose a novel deep learning recalibration module which shows significant gain in diagnosing celiac disease. In this recalibration module, the block-wise recalibration component is newly employed to capture the most salient feature in the local channel feature map. This learning module was embedded into ResNet50, Inception-v3 to diagnose celiac disease using a 10-time 10-fold cross-validation based upon analysis of VCE images. In addition, we employed model weights to extract feature points from training and test samples before the last fully connected layer, and then input to a support vector machine (SVM), k-nearest neighbor (KNN), and linear discriminant analysis (LDA) for differentiating celiac disease images from heathy controls. RESULTS: Overall, the accuracy, sensitivity and specificity of the 10-time 10-fold cross-validation were 95.94%, 97.20% and 95.63%, respectively. CONCLUSIONS: A novel deep learning recalibration module, with global response and local salient factors is proposed, and it has a high potential for utilizing deep learning networks to diagnose celiac disease using VCE images.
Authors: Tiago Ribeiro; Miguel Mascarenhas Saraiva; João P S Ferreira; Hélder Cardoso; João Afonso; Patrícia Andrade; Marco Parente; Renato Natal Jorge; Guilherme Macedo Journal: Ann Gastroenterol Date: 2021-07-02
Authors: Miguel Mascarenhas; Tiago Ribeiro; João Afonso; João P S Ferreira; Hélder Cardoso; Patrícia Andrade; Marco P L Parente; Renato N Jorge; Miguel Mascarenhas Saraiva; Guilherme Macedo Journal: Endosc Int Open Date: 2022-02-16