Enhancing Vision Transformer Performance with Rotation Based Augmentation for Classifying Images of Colon Cancer Pathology
Background: In medical imaging, classifying images of colon cancer pathology is still an essential challenge, especially for facilitating early diagnosis and successful intervention. Recently, Vision Transformer (ViT) models have demonstrated great promise for a variety of computer vision tasks, in...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universitas Nusantara PGRI Kediri
2025-07-01
|
| Series: | Intensif: Jurnal Ilmiah Penelitian Teknologi dan Penerapan Sistem Informasi |
| Subjects: | |
| Online Access: | https://ojs.unpkediri.ac.id/index.php/intensif/article/view/24918 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849319580405596160 |
|---|---|
| author | Rudy Eko Prasetya M. Arief Soeleman Farrikh Al Zami Affandy Affandy Aris Marjuni Mohammad Iqbal Saryuddin Assaqty |
| author_facet | Rudy Eko Prasetya M. Arief Soeleman Farrikh Al Zami Affandy Affandy Aris Marjuni Mohammad Iqbal Saryuddin Assaqty |
| author_sort | Rudy Eko Prasetya |
| collection | DOAJ |
| description |
Background: In medical imaging, classifying images of colon cancer pathology is still an essential challenge, especially for facilitating early diagnosis and successful intervention. Recently, Vision Transformer (ViT) models have demonstrated great promise for a variety of computer vision tasks, including the classification of medical images. However, the lack of annotated medical datasets and the intrinsic unpredictability of histopathology pictures sometimes restrict their performance. Objective: This study aims to enhance the performance of ViT models in colon cancer pathology classification by introducing a targeted data augmentation strategy, with a particular focus on rotation-based augmentation. Methods: We proposed a data augmentation pipeline that uses controlled changes to improve the number and diversity of training data. Like Rotation, Flip and Geometry are emphasized to replicate the real-world tissue orientation variations that are frequently seen in colon pathology slides. 10,000 JPEG pictures of colon cancer pathology, each with a resolution of 768 x 768 pixels, are used to train the models. We use models trained with and without the suggested augmentation pipeline to compare ViT performance across accuracy, sensitivity, and specificity in order to assess the impact of augmentation. Results: According to study results, rotation-based augmentation enhances ViT performance, achieving up to 99.30% accuracy and 99.50% sensitivity while preserving training times. In real-world pathology settings, where slide orientation varies greatly and can affect categorization consistency, these enhancements are especially pertinent. Conclusion: The proposed rotation-centric data augmentation technique enhances the performance of the ViT model in the classification of images showing colon cancer pathology.
|
| format | Article |
| id | doaj-art-eb06a389f8f64ba895bca132dce82f98 |
| institution | Kabale University |
| issn | 2580-409X 2549-6824 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Universitas Nusantara PGRI Kediri |
| record_format | Article |
| series | Intensif: Jurnal Ilmiah Penelitian Teknologi dan Penerapan Sistem Informasi |
| spelling | doaj-art-eb06a389f8f64ba895bca132dce82f982025-08-20T03:50:22ZengUniversitas Nusantara PGRI KediriIntensif: Jurnal Ilmiah Penelitian Teknologi dan Penerapan Sistem Informasi2580-409X2549-68242025-07-019210.29407/intensif.v9i2.24918Enhancing Vision Transformer Performance with Rotation Based Augmentation for Classifying Images of Colon Cancer PathologyRudy Eko Prasetya0https://orcid.org/0009-0001-5303-0890M. Arief Soeleman1https://orcid.org/0000-0001-6099-7023Farrikh Al Zami2https://orcid.org/0000-0003-2669-3864Affandy Affandy3https://orcid.org/0000-0003-1897-8261Aris Marjuni4https://orcid.org/0000-0002-4072-3081Mohammad Iqbal Saryuddin Assaqty5https://orcid.org/0000-0001-7274-6299Unuversitas Dian NuswantoroUniversitas Dian Nuswantoro SemarangUniversitas Dian Nuswantoro SemarangUniversitas Dian Nuswantoro SemarangUniversitas Dian Nuswantoro SemarangSouth China University of Technology Background: In medical imaging, classifying images of colon cancer pathology is still an essential challenge, especially for facilitating early diagnosis and successful intervention. Recently, Vision Transformer (ViT) models have demonstrated great promise for a variety of computer vision tasks, including the classification of medical images. However, the lack of annotated medical datasets and the intrinsic unpredictability of histopathology pictures sometimes restrict their performance. Objective: This study aims to enhance the performance of ViT models in colon cancer pathology classification by introducing a targeted data augmentation strategy, with a particular focus on rotation-based augmentation. Methods: We proposed a data augmentation pipeline that uses controlled changes to improve the number and diversity of training data. Like Rotation, Flip and Geometry are emphasized to replicate the real-world tissue orientation variations that are frequently seen in colon pathology slides. 10,000 JPEG pictures of colon cancer pathology, each with a resolution of 768 x 768 pixels, are used to train the models. We use models trained with and without the suggested augmentation pipeline to compare ViT performance across accuracy, sensitivity, and specificity in order to assess the impact of augmentation. Results: According to study results, rotation-based augmentation enhances ViT performance, achieving up to 99.30% accuracy and 99.50% sensitivity while preserving training times. In real-world pathology settings, where slide orientation varies greatly and can affect categorization consistency, these enhancements are especially pertinent. Conclusion: The proposed rotation-centric data augmentation technique enhances the performance of the ViT model in the classification of images showing colon cancer pathology. https://ojs.unpkediri.ac.id/index.php/intensif/article/view/24918Vision TransformerData AugmentationImages ClassificationColon Cancer |
| spellingShingle | Rudy Eko Prasetya M. Arief Soeleman Farrikh Al Zami Affandy Affandy Aris Marjuni Mohammad Iqbal Saryuddin Assaqty Enhancing Vision Transformer Performance with Rotation Based Augmentation for Classifying Images of Colon Cancer Pathology Intensif: Jurnal Ilmiah Penelitian Teknologi dan Penerapan Sistem Informasi Vision Transformer Data Augmentation Images Classification Colon Cancer |
| title | Enhancing Vision Transformer Performance with Rotation Based Augmentation for Classifying Images of Colon Cancer Pathology |
| title_full | Enhancing Vision Transformer Performance with Rotation Based Augmentation for Classifying Images of Colon Cancer Pathology |
| title_fullStr | Enhancing Vision Transformer Performance with Rotation Based Augmentation for Classifying Images of Colon Cancer Pathology |
| title_full_unstemmed | Enhancing Vision Transformer Performance with Rotation Based Augmentation for Classifying Images of Colon Cancer Pathology |
| title_short | Enhancing Vision Transformer Performance with Rotation Based Augmentation for Classifying Images of Colon Cancer Pathology |
| title_sort | enhancing vision transformer performance with rotation based augmentation for classifying images of colon cancer pathology |
| topic | Vision Transformer Data Augmentation Images Classification Colon Cancer |
| url | https://ojs.unpkediri.ac.id/index.php/intensif/article/view/24918 |
| work_keys_str_mv | AT rudyekoprasetya enhancingvisiontransformerperformancewithrotationbasedaugmentationforclassifyingimagesofcoloncancerpathology AT mariefsoeleman enhancingvisiontransformerperformancewithrotationbasedaugmentationforclassifyingimagesofcoloncancerpathology AT farrikhalzami enhancingvisiontransformerperformancewithrotationbasedaugmentationforclassifyingimagesofcoloncancerpathology AT affandyaffandy enhancingvisiontransformerperformancewithrotationbasedaugmentationforclassifyingimagesofcoloncancerpathology AT arismarjuni enhancingvisiontransformerperformancewithrotationbasedaugmentationforclassifyingimagesofcoloncancerpathology AT mohammadiqbalsaryuddinassaqty enhancingvisiontransformerperformancewithrotationbasedaugmentationforclassifyingimagesofcoloncancerpathology |