International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
|
Volume 186 - Issue 13 |
Published: March 2024 |
Authors: Esam Mohammed Othman |
![]() |
Esam Mohammed Othman . Breast Cancer Multi-Class Classification Using ViT Model. International Journal of Computer Applications. 186, 13 (March 2024), 13-18. DOI=10.5120/ijca2024923504
@article{ 10.5120/ijca2024923504, author = { Esam Mohammed Othman }, title = { Breast Cancer Multi-Class Classification Using ViT Model }, journal = { International Journal of Computer Applications }, year = { 2024 }, volume = { 186 }, number = { 13 }, pages = { 13-18 }, doi = { 10.5120/ijca2024923504 }, publisher = { Foundation of Computer Science (FCS), NY, USA } }
%0 Journal Article %D 2024 %A Esam Mohammed Othman %T Breast Cancer Multi-Class Classification Using ViT Model%T %J International Journal of Computer Applications %V 186 %N 13 %P 13-18 %R 10.5120/ijca2024923504 %I Foundation of Computer Science (FCS), NY, USA
Breast cancer ranks as the most prevalent form of cancer among women worldwide, underscoring the importance of early detection for enhancing treatment success rates. The ability to accurately differentiate between malignant (aggressive) and benign breast tumors is crucial for determining appropriate treatment strategies. This research introduces a novel methodology leveraging Transformer models for the task of breast cancer image classification. Utilizing a Vision Transformer (ViT) pre-trained across a broad array of domains, this approach incorporates an ensemble of densely connected network layers specifically refined for a dataset dedicated to breast cancer imagery. The performance of this innovative model was rigorously evaluated against a benchmark dataset, demonstrating superior classification capabilities with remarkable accuracy levels—97.5% in binary categorizations and 94% in multi-class scenarios. The findings from this study underscore the potential of employing advanced Transformer models in the precise classification of breast tumors, thereby contributing to the advancement of diagnostic techniques in oncology.