0% Complete
فارسی
Home
/
سی و دومین کنفرانس ملی و دهمین کنفرانس بین المللی مهندسی زیست پزشکی ایران
Benchmarking nnU-Net vs. Custom 3D U-Net for Kidney Tumor Segmentation: A Controlled Study on KiTS19 Dataset
Authors :
Ariya Soleimany
1
Masoud Noroozi
2
Mohammad Saber Azimi
3
Alireza Karimian
4
Jafar Majidpour
5
Hossein Arabi
6
1- University of Isfahan
2- University of Isfahan
3- Shahid Beheshti University
4- University of Isfahan
5- University of Raparin
6- Geneva University Hospital
Keywords :
deep learning،medical image segmentation،3D U-Net،nnU-Net،kidney tumor
Abstract :
This study presents a controlled comparison between nnU-Net and a custom three-dimensional U-Net architecture for kidney and tumor segmentation using the KiTS19 dataset. Both models were trained under identical conditions, using the same preprocessing, data augmentation, and training protocols to isolate the effects of architecture and optimization strategies. Evaluation was conducted on 210 labeled cases, with 140 used for training and 70 for testing, using the Dice similarity coefficient and intersection over union metrics. The nnU-Net achieved Dice scores of 0.9683 for kidney and 0.8166 for tumor, while the custom U-Net obtained scores of 0.9450 and 0.6296, respectively. Similarly, intersection over union scores were 0.9390 and 0.7176 for nnU-Net, compared to 0.8997 and 0.5167 for the custom model. A non-parametric Wilcoxon signed-rank test confirmed that these differences were statistically significant, with a p-value that was extremely small and very close to zero. These findings highlight the role of deep supervision, architectural scaling, and automated configuration in enhancing segmentation accuracy for challenging medical imaging tasks.
Papers List
List of archived papers
Fast Reflection-Mode Ultrasound Computed Tomography Versus Conventional Pulse-Echo Technique
Elnaz Rostami Siahpoush - Haniye Fathi - Zahra Kavehvash
تاثیر قدرت مدیر عامل بر کیفیت گزارشگری مالی
یعقوب پور کریم - میلاد حبیب اللهی - ابوالفضل بخشی قیسناب
راهکارهای عملی برای اجرای موفق پروژههای هوش مصنوعی در ایران
ملینا عبدلی
The role of artificial intelligence in linking data-driven human resources with sustainable business branding
Baharak Yadegar Jamshidi - Hossein Bodaghi Khajeh Noubar
Biomechanical Contrast Between Native and Decellularized Triple-Negative Breast Tumors in Mice
Mohammad Javad Farjam - Saman Asadi - Ashkan Azimzadeh - Saeid Amanpour - AbdolMohammad Kajbafzadeh - Mohammad Ali Nazari
Prediction of cardiac arrhythmia via an improved hierarchical fused fuzzy deep learning
Arman Daliri - Nora Mahdavi
واقعاً چه چیزی در جلوگیری از تقلب علیه سازمان ها مؤثر است که تصمیم گیرندگان باید بدانند؟
شبنم بالازاده قره باغی - سعید مصدق - اسماعیل عبادی
Inverse Dynamics Analysis of the Crutch-Assisted Gait with a Lower-Limb Robotic Exoskeleton
Negin Nasirian - Milad Hosseini - Reza Norouzzadeh - Saeed Behzadipour
Neural Correlates of Reward and Punishment Processing During Gambling-Based Decision-Making: A Simultaneous EEG-fMRI Study
Elias Ebrahimzadeh - Amin Mohammad Mohammadi - Ahmad Hammoud - Lila Rajabion - Hamid Soltanian-Zadeh
Synthesis and Swelling Behavior of pH-Sensitive Chitosan/Polyvinylpyrrolidone Hydrogels for Drug Delivery Applications
Shaghayegh Zameni Nir - Hanieh Shokrkar - Niloofar Nasirpour
more
Samin Hamayesh - Version 42.5.2