0% Complete
English
صفحه اصلی
/
سی و دومین کنفرانس ملی و دهمین کنفرانس بین المللی مهندسی زیست پزشکی ایران
Leveraging Normal White Matter Hyperintensity Context for Enhanced Pathological Segmentation via Multi-Class Deep Learning
نویسندگان :
Mahdi Bashiri Bawil
1
Mousa Shamsi
2
Ali Fahmi Jafargholkhanloo
3
Abolhassan Shakeri Bavil
4
1- Tabriz University of Technology (Sahand)
2- Tabriz University of Technology (Sahand)
3- University of Mohaghegh Ardabili
4- Department of Radiology, Imam Reza Hospital Tabriz University of Medical Sciences Tabriz, Iran
کلمات کلیدی :
White matter hyperintensities (WMH)،deep learning،medical image segmentation،FLAIR MRI،multi-class classification،U-Net،pathological segmentation،neuroimaging
چکیده :
White matter hyperintensities (WMHs) on FLAIR MRI are critical indicators of cerebrovascular dysfunction associated with elevated risks of stroke, dementia, and death. Current automated segmentation methods suffer from false positive detection in periventricular regions, failing to distinguish normal or aging-related hyperintensities from pathologically significant lesions, which reduces clinical applicability and diagnostic accuracy. This study investigates whether training deep learning models to explicitly differentiate between normal and abnormal WMH improves pathological WMH segmentation performance compared to traditional binary approaches. Four state-of-the-art architectures (U-Net, Attention U-Net, DeepLabV3Plus, Trans-U-Net) were evaluated across two training scenarios using 1,974 FLAIR images from 100 MS patients with expert-annotated ground truths. Scenario 1 employed binary training (background vs abnormal WMH), while Scenario 2 utilized three-class training (background, normal WMH, abnormal WMH). Statistical analysis included paired t-tests and Cohen's d effect size calculations. U-Net achieved the most substantial improvement in Scenario 2 with 55.6% increase in Dice coefficient (0.693 vs 0.443) and 131% precision enhancement (p < 0.0001, Cohen's d = 0.971). Traditional CNN-based architectures demonstrated larger effect sizes than transformer-based models. The three-class training approach significantly enhances pathological WMH segmentation while maintaining clinical feasibility, providing a validated framework for improving automated neuroimaging tools' diagnostic utility.
لیست مقالات
لیست مقالات بایگانی شده
بررسی تاثیر انعطاف پذیری مالی بر رابطه بین مدیریت ریسک و اجتناب مالیاتی
علیرضا عظیمی ثانی - رضا پورتقی
GPU-Accelerated GRAPPA: A Fast Implementation Using PyTorch for MRI Reconstruction
Mehrdad Anvari-Fard - Mahdi Bazargani - Mohammad Javad Heidari - Hamid Soltanian-Zadeh
Alterations of Brain Activation Maps in Adults with ADHD During Risk-Related Decision-Making Evidence from the Balloon Analogue Risk Task
Bahar Kermani - Mahdi Mirzaee Barzegar - Alireza Shirazinodeh
Conductive Hydrogels in Biomedical Engineering: Current Status and Challenges
Elham Amiraslani - Zahra Mohammadi
The Adaptive Approach of Ensemble Deep Learning Model in OCT Image Classification
Hamed Aghapanah Roudsari - Ali Ghaderian - Mrteza Choubin
ارائه مدل E-UNETR2D جهت قطعه بندی عروق کرونر از روی تصاویر سی تی آنژیوگرافی
مصطفی رجب زاده - فواد قادری - حمیدرضا پورعلی اکبر - نصرالله مقدم چرکری
تحول دیجیتال: مروری بر ادبیات سیستماتیک و نقش پایداری
رعنا شهدآور - بهزاد صادق تمیز - اکبر حکمت نژاد - یاسر بختیاری
Parkinson’s Disease Classification Using EEG and a Hybrid EEGNet–LSTM Architecture
Pouya Taghipour Langrodi - Amirsadra Khodadadi - Ali Sadat Modaresi - Mohammad Ahadzadeh - Mostafa Rostami - Sadegh Madadi
یشبینی فرار مالیاتی مؤدیان حقوقی با تاکید بر مولفههای اقتصادی، مؤدیان و حسابرسان مالیاتی؛ با تکیه بر هوش مصنوعی
حسین بوذری
CRAFT-Flow: Cross-Attentional Refinement for Robust Optical Flow Estimation in Cardiac MRI via Deep Learning
Hamed Aghapanah Roudsari - Reza Ashiri Gudarzi - Morteza Choubin
بیشتر
ثمین همایش، سامانه مدیریت کنفرانس ها و جشنواره ها - نگارش 42.5.2