Convolutional neural network classification of ultrasound images by liver fibrosis stages based on echo-envelope statistics
Introduction: Assessing the stage of liver fibrosis during the diagnosis and follow-up of patients with diffuse liver disease is crucial. The tissue structure in the fibrotic liver is reflected in the texture and contrast of an ultrasound image, with the pixel brightness indicating the intensity of...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-06-01
|
| Series: | Frontiers in Physics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphy.2023.1164622/full |
| _version_ | 1827914857304817664 |
|---|---|
| author | Shinnosuke Hirata Akiho Isshiki Dar-In Tai Po-Hsiang Tsui Kenji Yoshida Tadashi Yamaguchi |
| author_facet | Shinnosuke Hirata Akiho Isshiki Dar-In Tai Po-Hsiang Tsui Kenji Yoshida Tadashi Yamaguchi |
| author_sort | Shinnosuke Hirata |
| collection | DOAJ |
| description | Introduction: Assessing the stage of liver fibrosis during the diagnosis and follow-up of patients with diffuse liver disease is crucial. The tissue structure in the fibrotic liver is reflected in the texture and contrast of an ultrasound image, with the pixel brightness indicating the intensity of the echo envelope. Therefore, the progression of liver fibrosis can be evaluated non-invasively by analyzing ultrasound images.Methods: A convolutional-neural-network (CNN) classification of ultrasound images was applied to estimate liver fibrosis. In this study, the colorization of the ultrasound images using echo-envelope statistics that correspond to the features of the images is proposed to improve the accuracy of CNN classification. In the proposed method, the ultrasound image is modulated by the 3rd- and 4th-order moments of pixel brightness. The two modulated images and the original image were then synthesized into a color image of RGB representation.Results and Discussion: The colorized ultrasound images were classified via transfer learning of VGG-16 to evaluate the effect of colorization. Of the 80 ultrasound images with liver fibrosis stages F1–F4, 38 images were accurately classified by the CNN using the original ultrasound images, whereas 47 images were classified by the proposed method. |
| first_indexed | 2024-03-13T02:51:13Z |
| format | Article |
| id | doaj.art-216f57d1730f4538967a1e4707a4dd2c |
| institution | Directory Open Access Journal |
| issn | 2296-424X |
| language | English |
| last_indexed | 2024-03-13T02:51:13Z |
| publishDate | 2023-06-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physics |
| spelling | doaj.art-216f57d1730f4538967a1e4707a4dd2c2023-06-28T11:48:51ZengFrontiers Media S.A.Frontiers in Physics2296-424X2023-06-011110.3389/fphy.2023.11646221164622Convolutional neural network classification of ultrasound images by liver fibrosis stages based on echo-envelope statisticsShinnosuke Hirata0Akiho Isshiki1Dar-In Tai2Po-Hsiang Tsui3Kenji Yoshida4Tadashi Yamaguchi5Center for Frontier Medical Engineering, Chiba University, Chiba, JapanDepartment of Medical Engineering, Graduate School of Science and Engineering, Chiba University, Chiba, JapanDepartment of Gastroenterology and Hepatology, Chang Gung Memorial Hospital, Taoyuan, TaiwanDepartment of Medical Imaging and Radiological Sciences, College of Medicine, Chang Gung University, Taoyuan, TaiwanCenter for Frontier Medical Engineering, Chiba University, Chiba, JapanCenter for Frontier Medical Engineering, Chiba University, Chiba, JapanIntroduction: Assessing the stage of liver fibrosis during the diagnosis and follow-up of patients with diffuse liver disease is crucial. The tissue structure in the fibrotic liver is reflected in the texture and contrast of an ultrasound image, with the pixel brightness indicating the intensity of the echo envelope. Therefore, the progression of liver fibrosis can be evaluated non-invasively by analyzing ultrasound images.Methods: A convolutional-neural-network (CNN) classification of ultrasound images was applied to estimate liver fibrosis. In this study, the colorization of the ultrasound images using echo-envelope statistics that correspond to the features of the images is proposed to improve the accuracy of CNN classification. In the proposed method, the ultrasound image is modulated by the 3rd- and 4th-order moments of pixel brightness. The two modulated images and the original image were then synthesized into a color image of RGB representation.Results and Discussion: The colorized ultrasound images were classified via transfer learning of VGG-16 to evaluate the effect of colorization. Of the 80 ultrasound images with liver fibrosis stages F1–F4, 38 images were accurately classified by the CNN using the original ultrasound images, whereas 47 images were classified by the proposed method.https://www.frontiersin.org/articles/10.3389/fphy.2023.1164622/fulldiffuse liver diseaseliver fibrosisultrasound imagetexture analysisecho-envelope statisticsconvolutional neural network |
| spellingShingle | Shinnosuke Hirata Akiho Isshiki Dar-In Tai Po-Hsiang Tsui Kenji Yoshida Tadashi Yamaguchi Convolutional neural network classification of ultrasound images by liver fibrosis stages based on echo-envelope statistics Frontiers in Physics diffuse liver disease liver fibrosis ultrasound image texture analysis echo-envelope statistics convolutional neural network |
| title | Convolutional neural network classification of ultrasound images by liver fibrosis stages based on echo-envelope statistics |
| title_full | Convolutional neural network classification of ultrasound images by liver fibrosis stages based on echo-envelope statistics |
| title_fullStr | Convolutional neural network classification of ultrasound images by liver fibrosis stages based on echo-envelope statistics |
| title_full_unstemmed | Convolutional neural network classification of ultrasound images by liver fibrosis stages based on echo-envelope statistics |
| title_short | Convolutional neural network classification of ultrasound images by liver fibrosis stages based on echo-envelope statistics |
| title_sort | convolutional neural network classification of ultrasound images by liver fibrosis stages based on echo envelope statistics |
| topic | diffuse liver disease liver fibrosis ultrasound image texture analysis echo-envelope statistics convolutional neural network |
| url | https://www.frontiersin.org/articles/10.3389/fphy.2023.1164622/full |
| work_keys_str_mv | AT shinnosukehirata convolutionalneuralnetworkclassificationofultrasoundimagesbyliverfibrosisstagesbasedonechoenvelopestatistics AT akihoisshiki convolutionalneuralnetworkclassificationofultrasoundimagesbyliverfibrosisstagesbasedonechoenvelopestatistics AT darintai convolutionalneuralnetworkclassificationofultrasoundimagesbyliverfibrosisstagesbasedonechoenvelopestatistics AT pohsiangtsui convolutionalneuralnetworkclassificationofultrasoundimagesbyliverfibrosisstagesbasedonechoenvelopestatistics AT kenjiyoshida convolutionalneuralnetworkclassificationofultrasoundimagesbyliverfibrosisstagesbasedonechoenvelopestatistics AT tadashiyamaguchi convolutionalneuralnetworkclassificationofultrasoundimagesbyliverfibrosisstagesbasedonechoenvelopestatistics |