0% Complete
فارسی
Home
/
سی و دومین کنفرانس ملی و دهمین کنفرانس بین المللی مهندسی زیست پزشکی ایران
2D Residual U-Net for Accurate Lumbar Vertebrae Segmentation in MRI-Based Low Back Pain Diagnosis using the SPIDER Dataset
Authors :
Armita Rahimi Borgi
1
Abdollah Zohrabi
2
Ali Kazemi
3
Mostafa Abdolghaffar
4
Ramin Kordi
5
Parastoo Farnia
6
Alireza Ahmadian
7
1- Department of Medical Physics and Biomedical Engineering, Tehran University of Medical Sciences, Research Center of Biomedical Technology and Robotics (RCBTR), Advanced Medical Technologies and Equipment Institute (AMTEI), Imam Khomeini Hospital Complex, Tehran University of Medical Sciences, Tehran, Iran
2- Department of Mathematics (Computer Science), Sharif University of Technology, Tehran, Iran
3- Department of Medical Physics and Biomedical Engineering, Tehran University of Medical Sciences, Research Center of Biomedical Technology and Robotics (RCBTR), Advanced Medical Technologies and Equipment Institute (AMTEI), Imam Khomeini Hospital Complex, Tehran University of Medical Sciences, Tehran, Iran
4- Parseh Intelligent Surgical Systems Co., R&D Dept, Tehran, Iran
5- Department of Sports Medicine, School of Medicine, Tehran University of Medical Sciences, Tehran, Iran
6- Department of Medical Physics and Biomedical Engineering, Tehran University of Medical Sciences, Research Center for Intelligent Technologies in Medicine (RCITM), Advanced Medical Technologies and Equipment Institute (AMTEI), Tehran University of Medical Sciences, Tehran, Iran
7- Department of Medical Physics and Biomedical Engineering, Tehran University of Medical Sciences, Research Center of Biomedical Technology and Robotics (RCBTR), Advanced Medical Technologies and Equipment Institute (AMTEI), Imam Khomeini Hospital Complex, Tehran University of Medical Sciences, Tehran, Iran
Keywords :
Low Back Pain (LBP)،Magnetic Resonance Imaging (MRI)،Segmentation،Deep Learning،ResUNet
Abstract :
Low back pain (LBP) affects nearly 12% of the global population and remains a leading cause of disability. Precise visualization of lumbar vertebrae and intervertebral discs is critical for detecting pathological changes and guiding clinical interventions. Accurate segmentation of these structures is essential for reliable diagnosis and treatment planning. Compared to X-ray or CT scans, magnetic resonance imaging (MRI) provides superior soft-tissue contrast, making it the preferred modality for comprehensive spinal assessments. However, manual segmentation of vertebrae is labor-intensive and prone to inter-observer variability, while semi-automatic approaches are often time-consuming and lack robustness in accurately identifying vertebral anatomical structures, particularly when applied to low-quality or diverse clinical MRI data. In this study, we propose a 2D Residual U-Net (ResUNet) for vertebral segmentation on the SPIDER dataset. The pipeline includes reorientation, resolution standardization, and morphological mask refinement, along with a hybrid Dice–Binary Cross-Entropy loss to address class imbalance, particularly in non-vertebral structures. The proposed model achieved a Dice score of 0.945 and an IoU of 0.897, slightly surpassing a standard 2DU-Net with Dice = 0.940 and IoU=0.823. These results demonstrate that accurate 2D segmentation enables reliable 3D reconstruction, providing an efficient and clinically applicable solution for spinal analysis and LBP diagnosis.
Papers List
List of archived papers
Perfluorocarbon-Based Oxygenation Systems: From Foundational Principles to Revolutionary Applications in Cancer Therapy and Tissue Engineering
Gity Mirzaei - Zeinab Mazloumi - Ali Baradar Khoshfetrat
Document Clustering Using Deep Pre-trained Language Model Embeddings for Information Retrieval
Mahdi Mohammadiha - Mohammad Hassan Sadreddini - Morteza Mohammadi Zanjireh
آلیاژهای حافظهدار نیکل-تیتانیم در مهندسی پزشکی: نوآوریها، چالشها و کاربردهای پزشکی
مهدیه سلطانعلی پور - میلاد بدر - جعفر خلیل علافی
A Combined Time-Frequency and Common Spatial-Spectral Pattern Approach for EEG-Based Motor Imagery Classification
Reza Nejati - Hamed Danandeh Hesar
Robust Binary Differentiation of ALL vs. AML Using Deep Graph Convolutions
Mahsan Rahmani - Saeed Meshgini - Reza Afrouzian
بررسی تأثیر بالقوه فناوری بلاکچین بر کاهش مدیریت سود در شرکتهای بورسی: یک تحلیل مبتنی بر نگرش کارشناسان و متخصصین
ضرغام داداش زاده
تحلیلی بر آموزش حسابداری در ایران
محمد رضا پور علی - آرمان محجوب
عصر جدید مدلسازی بیومکانیکی با یادگیری ماشین آگاه از فیزیک
علی یعقوبیان - فائزه یعقوبیان
کاربرد هوش مصنوعی در بهینهسازی پردازش معاملات مالی: مدل های مفهومی و آیندهپژوهی
سجاد یوسفی - مریم پورنجف - لیلا حسین آبادی
افزایش پیش بینی بازار سهام از طریق هوش مصنوعی
سهیلا صمدی گلوجه - اسما حیدری پناه - زهرا علی لیواری - فاطمه خالقیان
more
Samin Hamayesh - Version 42.5.2