Citation

Yazid, Haniza and Abdul Aziz, Anis Amira and Mustafa, Nazahah and Abdul Rahim, Saufiah and Mat Som, Mohd Hanafi (2026) Impact of Image Pre-Processing and Optimizers on U-Net Segmentation of Prostate Cancer in MRI. International Journal on Robotics Automation and Sciences, 8 (1). p. 104. ISSN 2682-860X

Text 3.pdf - Published Version Restricted to Repository staff only Download (724kB)

Abstract

Prostate cancer remains among the most frequently diagnosed cancers in men and is a significant issue in achieving timely and accurate diagnosis. Magnetic Resonance Imaging (MRI) is widely used because it can capture high-resolution anatomical information; however, cancer regions must be manually segmented, which is time-consuming and prone to variability among experts. This study proposes an automated segmentation algorithm utilizing the U-Net deep learning model, combined with image preprocessing techniques, to address these deficiencies. Median filtering was applied to remove salt-and-pepper noise, followed by brightness enhancement to enhance the intensity contrast of the images. The pre-processed images were used to train a U-Net model for segmenting prostate cancer. The Dice Similarity Coefficient (DSC) metric was used to evaluate segmentation accuracy. Three optimizers, Adam, RMSprop, and Adagrad, were tested. All of them were trained between 10 and 100 epochs. The Adam optimizer achieved the highest segmentation performance at epoch 90, with a DSC value of 0.9907, while RMSprop and Adagrad yielded 0.9888 and 0.9655, respectively. Pre-processing raised the mean DSC from 0.8206 to 0.8733, confirming its impact on image quality enhancement. Overall, the proposed method demonstrates high accuracy and reliability, offering a practical solution to support radiologists in prostate cancer diagnosis and treatment planning.

Item Type:	Article
Uncontrolled Keywords:	Prostate cancer, Magnetic Resonance Imaging (MRI)
Subjects:	R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	09 Jul 2026 03:19
Last Modified:	09 Jul 2026 03:19
URII:	http://shdl.mmu.edu.my/id/eprint/16329

Downloads

Edit (login required)