Citation

Tan, Choo Hui and Lim, Wei Jie and Wan Ahmad, Wan Siti Halimatul Munirah and Wong, Lai Kuan and Ur Rehman, Zaka and Looi, Lai Meng and Cheah, Phaik Leng and Toh, Fa Yen and Ahmad Fauzi, Mohammad Faizal (2023) HER2-Sish Histopathology Image Classification Using Deep Neural Networks. In: 2023 IEEE International Conference on Image Processing (ICIP), 08-11 October 2023, Kuala Lumpur, Malaysia.

Full text not available from this repository.

Abstract

The status of the human epidermal growth factor receptor 2 (HER2) gene amplification is an important marker for assessing the efficacy of clinical treatments for breast cancer. This article discusses the application of deep learning to classify HER2-SISH (silver-enhanced in situ hybridization) pathological images and identifies their HER2/Chr17 amplification status. We used four pre-trained models for classifying the cases into either amplified or non-amplified: two models from the convolutional neural networks, CNNs (DenseNet, and MobileNet), and two transformer models (Vision Transformer, and Data-Efficient Image Transformers). Apart from these single models, we also built two ensemble models by concatenating the transformer and CNN architectures to observe their performances. A private dataset obtained from our collaborating hospital is used in this project, with several preprocessing techniques applied to the raw images prior to feeding the models. Promising results are reported with ViT emerged as the best performing model with a high accuracy of 87.48%, with 92.93% recall in detecting amplified HER2-SISH samples.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	breast cancer, digital pathology, deep learning, SISH, HER2, vision transformer, CNN
Subjects:	Q Science > QC Physics > QC350-467 Optics. Light
Divisions:	Faculty of Computing and Informatics (FCI) Faculty of Engineering (FOE)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	23 Feb 2024 04:13
Last Modified:	23 Feb 2024 04:13
URII:	http://shdl.mmu.edu.my/id/eprint/12129

Downloads

Edit (login required)