Citation
Touhami, Meriem and Ur Rehman, Zaka and Ahmad Fauzi, Mohammad Faizal and Mansor, Sarina (2024) Comparison of Conventional and U-Net Based Histopathology Image Enhancement. In: 2024 IEEE 8th International Conference on Signal and Image Processing Applications (ICSIPA), 03-05 September 2024, Kuala Lumpur, Malaysia.
Text
b.pdf - Published Version Restricted to Repository staff only Download (2MB) |
Abstract
The high resolution histopathology whole slide images (WSI) is important for making precise diagnosis and prognosis of cancer regions of digitized scanned glass slides. However, scanning high-quality WSI can be difficult due to the extensive scanning time and the large file size generated. To address these issues, the images taken by low quality microscopes can be enhanced by computational techniques. This paper conducted a research in order to present a comparative study between a conventional image enhancement method and a deep learning method for the enhancement of histopathology images. The used methods provide clearer and more detailed visual information, the findings, based on visual results and quantitative metrics, demonstrate the superiority of the U-Net method over histogram equalization. While histogram equalization effectively improves image contrast, it often leads to over-enhancement and the loss of critical details, resulting in less reliable outcomes. In contrast, the U-Net method excels in preserving complex details and providing a more accurate and reliable enhancement. Consequently, we conclude that deep learning methods, such as U-Net, show great promise in improving the quality of histopathology images, thereby supporting more accurate diagnoses.
Item Type: | Conference or Workshop Item (Paper) |
---|---|
Uncontrolled Keywords: | Deep learning, Histograms, Visualization, Accuracy, Image resolution, Histopathology, Reliability |
Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800-8360 Electronics > TK8300-8360 Photoelectronic devices (General) |
Divisions: | Faculty of Engineering (FOE) |
Depositing User: | Ms Nurul Iqtiani Ahmad |
Date Deposited: | 04 Nov 2024 01:24 |
Last Modified: | 04 Nov 2024 01:24 |
URII: | http://shdl.mmu.edu.my/id/eprint/13079 |
Downloads
Downloads per month over past year
Edit (login required) |