Accepted Articles
Back to the articles list |
Back to browse issues page
Zahra Zolghadr1 
, Seyed Amir Hossein Batouli2 , Hamid Alavimajd1 , Lida Shafaghi3 , Yadollah Mehrabi * 4
, Seyed Amir Hossein Batouli2 , Hamid Alavimajd1 , Lida Shafaghi3 , Yadollah Mehrabi * 4
1- Department of Biostatistics, School of Allied Medical Sciences, Shahid Beheshti University of Medical Sciences
2- Department of Neuroscience and Addiction Studies, School of Advanced Technologies in Medicine, Tehran University of Medical Sciences, Tehran, Iran.
3- Department of Neuroscience and Addiction Studies, School of Advanced Technologies in Medicine, Tehran University of Medical Sciences, Tehran, Iran
4- School of Public Health and Safety, Shahid Beheshti University of Medical Sciences, Tehran, Iran
2- Department of Neuroscience and Addiction Studies, School of Advanced Technologies in Medicine, Tehran University of Medical Sciences, Tehran, Iran.
3- Department of Neuroscience and Addiction Studies, School of Advanced Technologies in Medicine, Tehran University of Medical Sciences, Tehran, Iran
4- School of Public Health and Safety, Shahid Beheshti University of Medical Sciences, Tehran, Iran
Abstract:
Background: Neurodevelopmental disorders are a group of neuropsychiatric. Behavior-based diagnostic approaches are currently used in the clinical setting, but the overlapping features among the disorders vague the recognition and management of these disorders. Attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) have common characteristics throughout the hierarchy, from genes to symptoms. Designing a computational framework based on the neuroimaging findings could provide a discriminative tool for ultimate more efficient treatment. Machine learning approaches, specifically classification methods are among the most applied techniques to reach this goal.
Method: We applied a novel two-level multi-class DMDC algorithm to classify the functional neuroimaging data (Utilized Datasets: ADHD-200 and ABIDE) for the two classes of neurodevelopmental disorders (ASD and ADHD) or healthy participant, according to the calculated functional connectivity values (statistical temporal correlation).
Result: Regarding the healthy controls, our model yielded a total accuracy of 62%. Respectively, the model was 51% accurate for healthy subjects, 61% and 84% for autism spectrum disorder and ADHD. The SVM model provided the accuracy of 46% for the healthy control and ASD groups, ADHD group classification accuracy estimated to be 84%. These two models showed similar classification indices for ADHD group. However, the discrimination power was higher in class of autism spectrum disorder.
Conclusion: The currently applied method showed acceptable applications for classifying disorder and healthy conditions, compared to the more applied SVM method. The functional connections related to the cerebellum exhibited the discriminative power.
Method: We applied a novel two-level multi-class DMDC algorithm to classify the functional neuroimaging data (Utilized Datasets: ADHD-200 and ABIDE) for the two classes of neurodevelopmental disorders (ASD and ADHD) or healthy participant, according to the calculated functional connectivity values (statistical temporal correlation).
Result: Regarding the healthy controls, our model yielded a total accuracy of 62%. Respectively, the model was 51% accurate for healthy subjects, 61% and 84% for autism spectrum disorder and ADHD. The SVM model provided the accuracy of 46% for the healthy control and ASD groups, ADHD group classification accuracy estimated to be 84%. These two models showed similar classification indices for ADHD group. However, the discrimination power was higher in class of autism spectrum disorder.
Conclusion: The currently applied method showed acceptable applications for classifying disorder and healthy conditions, compared to the more applied SVM method. The functional connections related to the cerebellum exhibited the discriminative power.
Keywords: Functional Connectivity, DMDC, fMRI, Classification, Autism, ADHD, High-Dimensional Low Sample Size, ADHD-200, ABIDE
Type of Study: Original |
Subject:
Computational Neuroscience
Received: 2022/06/11 | Accepted: 2022/12/25
Received: 2022/06/11 | Accepted: 2022/12/25
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
Contact Information
Basic and Clinical Neuroscience Journal (BCN)
Iran University of Medical Sciences, Shahid Hemmat Highway, Tehran, Iran.
Publisher Tel:
+9821 453 55 555; +9821 453 55 000 (EXT: 104)
URL: http://bcn.iums.ac.ir/
Publisher Email: info
nirp.ir
