Publications by Farhad Abedinzadeh Torghabeh

Publications by authors named "Farhad Abedinzadeh Torghabeh"

Neuroinformatics

October 2024

ADHD is a common neurobehavioral disorder in kids and teens that needs early detection, and EEG connectivity measures can help improve its diagnosis.
This study presents a new ADHD diagnostic approach using a combination of connectivity maps derived from EEG data and a specialized convolutional neural network (Att-CNN).
The proposed method achieved high performance metrics (accuracy of 98.88% and F1 Score of 98.30%) with the help of advanced optimizers, suggesting it could significantly enhance early diagnosis and treatment efficacy for ADHD.

Gait Posture

September 2024

* The study investigates two research questions: the effectiveness of analyzing limb interactions in the time-frequency domain for classifying NDDs and the use of color-coded images with deep learning models for the same purpose.
* Models like AlexNet and GoogLeNet showed high accuracy in classifying gait signals from patients with various NDDs, with AlexNet reaching up to 99.20% accuracy, indicating that the methodology can significantly enhance diagnosis and treatment strategies.

Med Biol Eng Comput

February 2024

Early diagnosis of autism spectrum disorder (ASD) is critical for patient rehabilitation, requiring advanced pattern recognition and modeling techniques.
The study utilizes scalogram images from electroencephalography signals combined with a two-level deep learning architecture to enhance classification accuracy.
Results from testing on a dataset of 34 ASD samples and 11 normal cases show high detection performance, achieving 99.50% accuracy with voice and 98.43% without, indicating the method's effectiveness and the influence of auditory factors on diagnosis.

Phys Eng Sci Med

December 2023

* The study uses brain connectivity measurements and six classifiers to differentiate between children with ADHD and those without, achieving an impressive accuracy of 99.174%.
* Two potential biomarkers related to brain connectivity patterns are proposed, with statistical validation indicating their significant difference, highlighting their promise for improving ADHD diagnosis and intervention strategies.