StomachNet: Optimal Deep Learning Features Fusion for Stomach Abnormalities Classification
A fully automated design is proposed in this work employing optimal deep learning features for classifying gastrointestinal infections. Here, three prominent infections– ulcer, bleeding, polyp and a healthy class are considered as class labels. In the initial stage, the contrast is improv...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9240940/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850209390797258752 |
|---|---|
| author | Muhammad Attique Khan Muhammad Shahzad Sarfraz Majed Alhaisoni Abdulaziz A. Albesher Shuihua Wang Imran Ashraf |
| author_facet | Muhammad Attique Khan Muhammad Shahzad Sarfraz Majed Alhaisoni Abdulaziz A. Albesher Shuihua Wang Imran Ashraf |
| author_sort | Muhammad Attique Khan |
| collection | DOAJ |
| description | A fully automated design is proposed in this work employing optimal deep learning features for classifying gastrointestinal infections. Here, three prominent infections– ulcer, bleeding, polyp and a healthy class are considered as class labels. In the initial stage, the contrast is improved by fusing bi-directional histogram equalization with top-hat filtering output. The resultant fusion images are then passed to ResNet101 pre-trained model and trained once again using deep transfer learning. However, there are challenges involved in extracting deep learning features including impertinent information and redundancy. To mitigate this problem, we took advantage of two metaheuristic algorithms– Enhanced Crow Search and Differential Evolution. These algorithms are implemented in parallel to obtain optimal feature vectors. Following this, a maximum correlation-based fusion approach is applied to fuse optimal vectors from the previous step to obtain an enhanced vector. This final vector is given as input to Extreme Learning Machine (ELM) classifier for final classification. The proposed method is evaluated on a combined database. It accomplished an accuracy of 99.46%, which shows significant improvement over preceding techniques and other neural network architectures. |
| format | Article |
| id | doaj-art-fb93ea21f415468d96bd7e5e6950fe48 |
| institution | OA Journals |
| issn | 2169-3536 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-fb93ea21f415468d96bd7e5e6950fe482025-08-20T02:10:01ZengIEEEIEEE Access2169-35362020-01-01819796919798110.1109/ACCESS.2020.30342179240940StomachNet: Optimal Deep Learning Features Fusion for Stomach Abnormalities ClassificationMuhammad Attique Khan0https://orcid.org/0000-0001-7058-0715Muhammad Shahzad Sarfraz1Majed Alhaisoni2https://orcid.org/0000-0002-0231-6899Abdulaziz A. Albesher3https://orcid.org/0000-0002-4879-9128Shuihua Wang4https://orcid.org/0000-0003-2238-6808Imran Ashraf5https://orcid.org/0000-0003-4480-2489Department of Computer Science, HITEC University, Taxila, PakistanDepartment of Computer Science, National University of Computer and Emerging Sciences at Chiniot–Faisalabad Campus, Chiniot, PakistanCollege of Computer Science and Engineering, University of Ha’il, Ha’il, Saudi ArabiaCollege of Computing and Informatics, Saudi Electronic University, Riyadh, Saudi ArabiaDepartment of Mathematics, University of Leicester, Leicester, U.K.Department of Computer Science, HITEC University, Taxila, PakistanA fully automated design is proposed in this work employing optimal deep learning features for classifying gastrointestinal infections. Here, three prominent infections– ulcer, bleeding, polyp and a healthy class are considered as class labels. In the initial stage, the contrast is improved by fusing bi-directional histogram equalization with top-hat filtering output. The resultant fusion images are then passed to ResNet101 pre-trained model and trained once again using deep transfer learning. However, there are challenges involved in extracting deep learning features including impertinent information and redundancy. To mitigate this problem, we took advantage of two metaheuristic algorithms– Enhanced Crow Search and Differential Evolution. These algorithms are implemented in parallel to obtain optimal feature vectors. Following this, a maximum correlation-based fusion approach is applied to fuse optimal vectors from the previous step to obtain an enhanced vector. This final vector is given as input to Extreme Learning Machine (ELM) classifier for final classification. The proposed method is evaluated on a combined database. It accomplished an accuracy of 99.46%, which shows significant improvement over preceding techniques and other neural network architectures.https://ieeexplore.ieee.org/document/9240940/Stomach infectionscontrast stretchingdeep learningoptimizationfusion |
| spellingShingle | Muhammad Attique Khan Muhammad Shahzad Sarfraz Majed Alhaisoni Abdulaziz A. Albesher Shuihua Wang Imran Ashraf StomachNet: Optimal Deep Learning Features Fusion for Stomach Abnormalities Classification IEEE Access Stomach infections contrast stretching deep learning optimization fusion |
| title | StomachNet: Optimal Deep Learning Features Fusion for Stomach Abnormalities Classification |
| title_full | StomachNet: Optimal Deep Learning Features Fusion for Stomach Abnormalities Classification |
| title_fullStr | StomachNet: Optimal Deep Learning Features Fusion for Stomach Abnormalities Classification |
| title_full_unstemmed | StomachNet: Optimal Deep Learning Features Fusion for Stomach Abnormalities Classification |
| title_short | StomachNet: Optimal Deep Learning Features Fusion for Stomach Abnormalities Classification |
| title_sort | stomachnet optimal deep learning features fusion for stomach abnormalities classification |
| topic | Stomach infections contrast stretching deep learning optimization fusion |
| url | https://ieeexplore.ieee.org/document/9240940/ |
| work_keys_str_mv | AT muhammadattiquekhan stomachnetoptimaldeeplearningfeaturesfusionforstomachabnormalitiesclassification AT muhammadshahzadsarfraz stomachnetoptimaldeeplearningfeaturesfusionforstomachabnormalitiesclassification AT majedalhaisoni stomachnetoptimaldeeplearningfeaturesfusionforstomachabnormalitiesclassification AT abdulazizaalbesher stomachnetoptimaldeeplearningfeaturesfusionforstomachabnormalitiesclassification AT shuihuawang stomachnetoptimaldeeplearningfeaturesfusionforstomachabnormalitiesclassification AT imranashraf stomachnetoptimaldeeplearningfeaturesfusionforstomachabnormalitiesclassification |