Please wait ...
0% Complete
Home
/
14th International Conference on Computer and Knowledge Engineering
Optimizing MR Image Registration for Accurate Brain Volume Measurement in Children with Autism Spectrum Disorder
Authors :
Shiva Sanati
1
Mahdi Saadatmand
2
1- Ph.D. in Computer Engineering, Ferdowsi University of Mashhad, Mashhad, Iran
2- Associate Professor in Electrical Engineering, Ferdowsi University of Mashhad
Keywords :
Magnetic Resonance Imaging (MRI)،Autism Spectrum Disorder (ASD)،Image Registration،Brain Volume Measurement،VBM Dartel،SVM Classifier
Abstract :
Autism Spectrum Disorder (ASD) encompasses a range of neurodevelopmental conditions, highlighting the need for early and accurate diagnosis. This study focuses on identifying regions of interest (ROIs) in the brain that differ significantly between individuals with ASD and typically developing subjects (TDS), aiming to classify these groups with high accuracy. We introduce novel algorithms for registering brain MR images with the ICBM152 digital atlas, employing both rigid (affine) and non-rigid transformations. The volumes of 116 gray matter regions were measured using the MNI-AAL atlas aligned with ICBM152, and the volume differences between autistic and typically developing subjects were analyzed. The study utilized MR images from the ABIDE database, including 26 typically developing and 26 autistic children aged 5 to 10 years. Our registration algorithms significantly outperformed established methods, achieving a cross-correlation coefficient (CCC) of 0.82 for the ASD group, compared to 0.65 using the SPM toolbox's Powell method. Non-rigid registration further enhanced the CCC to 0.88, surpassing the 0.83 obtained by SPM's DCT-based method, and reduced convergence time by 2.8 times. Comparison with the Dartel VBM method showed consistent regions of difference between ASD and TDS, with significant increases in gray matter volumes in the right calcarine, left precuneus, right thalamus, left posterior cingulate gyrus, and bilateral lingual gyri in the ASD group. These results align with the age-specific theory of autism and provide insights into the structural brain changes associated with ASD. Finally, our SVM-based classification method achieved an accuracy of 92.3%, with an F1-score of 90.1%, recall of 90.1%, and precision of 89.4%, demonstrating the effectiveness of the proposed approach in differentiating ASD from TDS based on brain features.
Papers List
List of archived papers
Mitochondrial Segmentation in Microscopy Images Using UNet-VGG19
Zerek Sediq Hossein - Rojiar Pir Mohammadiani - Saadat Izadi
A Review on Secure Data Storage and Data Sharing Technics in Blockchain-based IoT Healthcare Systems
Seyedeh Somayeh Fatemi Nasab - Davoud Bahrepour - Seyed Reza Kamel Tabbakh
An Energy-efficient Clustering Method based on Butterfly Optimization Algorithm by Considering the Criterion of Intra-cluster Distances in WSNs
Fariba Saghi Hadi S. Aghdasi
Ensemble-Based Fraud Detection: A Robust Approach Evaluated on IEEE-CIS
Fatemeh Moradi - Mehran Tarif - Mohammadhossein Homaei
Persian Legal Text Simplification Leveraging Transformer-Based Models
Mohammadreza Joneidi Jafari - Saedeh Tahery - Amirhossein Nikoofard
FedBrain-Distill: Communication-Efficient Federated Brain Tumor Classification Using Ensemble Knowledge Distillation on Non-IID Data
Rasoul Jafari Gohari - Laya Aliahmadipour - Ezat Valipour
Binary Classification of Capuchin Bird Calls via Spectrogram-Enhanced Frequency-Aware Convolutional Neural Networks
Samad Najjar-Ghabel - Shamim Yousefi - Reza Danandeh Bileh Savar
Capturing Local and Global Features in Medical Images by Using Ensemble CNN-Transformer
Javad Mirzapour Kaleybar - Hooman Saadat - Hooman Khaloo
Facial Mask Wearing Condition Detection Using SSD MobileNetV2
Amirhossein Tighkhorshid - Yasamin Borhani - Javad Khoramdel - Esmaeil Najafi
Semi-automatic Detection of Persian Stopwords using FastText Library
Mohammad Dehghani - Mohammad Manthouri
more
Samin Hamayesh - Version 46.0.2