Cognitive Neurodynamics - Special Issue Call for Paper

Special issue title

Advances in Deep Convolutional Neural Networks for Adult ADHD Assessment from a Neuropsychological Perspective

Guest Editors

Dr. C .Chandru Vignesh, Department of Computer Science and Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, India
Dr. C.B. Sivaparthipan, Department of Computer Science and Engineering, Adhiyamaan College of Engineering, Tamil Nadu, India
Dr. Adhiyaman Manickam, Department of Computer Science, University of Moncton, Canada

About the special issue

ADHD (attention-deficit/hyperactive disorder) is a neurodevelopmental disorder beginning in childhood, characterized by differences in cognitive, emotional, and behavioral responses. Motor skills are significantly impaired in ADHD children, especially in the development of gross motor skills. ADHD is currently used to define the pattern of symptoms, not a specific disease entity. The clinical implications of the biological bases of ADHD have assessed brain structure and function, an important topic in this field. Since adult ADHD has a high level of comorbidity with other psychiatric disorders and substance abuse, assessing the presence of ADHD in adults allows for an additional diagnosis that can guide therapy for other conditions. Thus, the efficient diagnosis of ADHD from a neuropsychological perspective has become crucial. Deep convolutional neural networks are conceptual models that consist of multiple processing layers whose operations are determined by the connection patterns within and between their constituent layers. Although previous studies have identified certain network structures with high success rates for ADHD diagnosis, many of these studies relied on low sample sizes and exhibited limited generalizability.

Deep convolutional neural networks (CNNs) is a novel approach to assessing adult ADHD based on neuropsychological tests. While other approaches are available, they suffer from significant limitations with respect to constructing validity and test-retest reliability. In CNNs, the network is fully connected until the output neurons. Backpropagation often trains deep convolutional neural networks in a supervised learning framework. CNNs are able to adapt and learn contextual features based on raw scores of neuropsychological test scores of ADHD participants, but not based on normalized scores described in conventional scoring methodologies. Thus, it is possible for CNNs to better identify underlying characteristics that are specific to ADHD. For example,  those who were diagnosed with ADHD during adolescence showed abnormalities on mathematical learning tasks; those with executive function deficits tended to be more attentive than those without such deficits; and some tests have surprisingly poor validity for the diagnosis of ADHD because their scores do not discriminate well between ADHD participants, particularly in the areas of processing efficiency and variability.

This special issue explores recent advances in deep convolutional neural networks (CNNs) for assessing adult ADHD, based on a neuropsychological perspective. It collectively provides the rationale for using CNNs in the field of ADHD assessment. Researchers interested in understanding how the use of deep convolutional neural networks can help bridge between psychology and neuroscience by presenting their novel and innovative contributions.

Topics of interest for the special issue:

• Advances in deep convolutional neural networks for adult ADHD assessment from a neuropsychological perspective
• Deep learning assisted clinical neuropsychology solutions for adult ADHD assessment
• Deep learning based neuroimaging approaches for ADHD interventions at an earlier stage among adults
• Assessing brain image and behavior assessment with deep convolutional neural networks for adult ADHD assessment and diagnosis
• Heterogeneous neural development in adults with deep convolutional neural networks
• Advances in deep recurrent neural networks for EEG signal processing for ADHD assessment
• New research perspectives in deep learning for brain neuroimaging in diagnosis of ADHD assessment
• Explainable deep learning for EEG data acquisition and processing
• Innovations in neural network architectures for effective assessment of ADHD in adults
• Advances in convolutional neural networks for effective classification of ADHD adult patients and early diagnosis.

Important Dates

Submission Deadline: 30 December 2022
Authors Notification: 25 March 2023
Revised Papers Due: 18 June 2023
Final notification: 26 September 2023