Ha Min Son

Introduction: Attention-Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder where hyperactivity often manifests as fidgeting, a non-goal-directed motoric action. Many studies demonstrate fidgeting varies under different conditions as a self-regulating mechanism for attention and alertness during cognitively demanding tasks. Fidgeting has also been associated with reaction time variability. However, a lack of standard variables to define and quantify fidgeting can lead to discrepancies in data and interpretability issues across studies. Furthermore, little is known about fidgeting in adults with ADHD compared to youth. This study aims to design a framework to quantify meaningful fidgeting variables and to apply them to test the relation between fidgeting and performance on a cognitive task, the Flanker, in adults with ADHD.Method: Our study included 70 adult participants diagnosed with ADHD, aged 18–50 years (30.5 ± 7.2 years). Screening included a structured clinical interview, childhood, current self and current observer ratings of ADHD symptoms. Actigraphy devices were attached to the left wrist and right ankle during completion of a cognitive control, attention task (the Flanker). Laboratory testing was subsequently completed on a single day. The relation between task performance, reaction time variability and fidgeting was examined.Results and Discussion: Our analysis revealed increased fidgeting during correct trials as defined by our new variables, consistent with previous observations. Furthermore, differences in fidgeting were observed between early and later trials while the percentage of correct trials were not significantly different. This suggests a relation between the role of fidgeting and sustaining attention. Participants with low reaction time variability, that is, those with more consistent reaction times, fidgeted more during later trials. This observation supports the theory that fidgeting aids arousal and improves sustained attention. Finally, a correlation analysis using ADHD-symptom rating scales validated the relevance of the fidget variables in relation to ADHD symptom severity. These findings suggest fidgeting may be a compensatory mechanism that aids in sustained attention for those with ADHD, although alternative explanations exist.Conclusion: Our study suggests that fidgeting may aid in sustained attention during the attention-demanding, cognitive control processes for adults with ADHD, with more fidgeting observed during correct trials and among participants with lower reaction time variability. Furthermore, the newly defined fidget variables were validated through a significant correlation with ADHD rating scales. By sharing our implementation of fidget variables, we hope to standardize and encourage further quantitative research into the role of fidgeting in ADHD.

Federated learning is a promising framework to train neural networks with widely distributed data. However, performance degrades heavily with heterogeneously distributed data. Recent work has shown this is due to the final layer of the network being most prone to local bias, some finding success freezing the final layer as an orthogonal classifier. We investigate the training dynamics of the classifier by applying SVD to the weights motivated by the observation that freezing weights results in constant singular values. We find that there are differences when training in IID and non-IID settings. Based on this finding, we introduce two regularization terms for local training to continuously emulate IID settings: (1) variance in the dimension-wise probability distribution of the classifier and (2) hyperspherical uniformity of representations of the encoder. These regularizations promote local models to act as if it were in an IID setting regardless of the local data distribution, thus off-setting proneness to bias while being flexible to the data. On extensive experiments in both label-shift and feature-shift settings, we verify that our method achieves highest performance by a large margin especially in highly non-IID cases in addition to being scalable to larger models and datasets.

Introduction: Diagnosis of Attention Deficit/Hyperactivity Disorder (ADHD) is based on clinical evaluation of symptoms by a psychiatrist, referencing results of psychological tests. When diagnosing ADHD, the child’s behavior and functionality in real-life situations are critical components. However, direct observation by a clinician is often not feasible in practice. Therefore, such information is typically gathered from primary caregivers or teachers, which can introduce subjective elements. To overcome these limitations, we developed AttnKare-D, an innovative digital diagnostic tool that could analyze children’s behavioral data in Virtual Reality using Artificial Intelligence. The purpose of this study was to explore the utility and safety of AttnKare-D for clinical application.Method: A total of 21 children aged between 6 and 12 years were recruited for this study. Among them, 15 were children diagnosed with ADHD, 5 were part of a normal control group, and 1 child was excluded due to withdrawal of consent. Psychological assessments, including K-WISC, Conners CPT, K-ARS, and K-CBCL, were conducted for participants and their primary caregivers. Diagnoses of ADHD were confirmed by child and adolescent psychiatrists based on comprehensive face-to-face evaluations and results of psychological assessments. Participants underwent VR diagnostic assessment by performing various cognitive and behavioral tasks in a VR environment. Collected data were analyzed using an AI model to assess ADHD diagnosis and the severity of symptoms.Results: AttnKare-D demonstrated diagnostic performance with an AUC of 0.893 when compared to diagnoses made by child and adolescent psychiatrist, showing a sensitivity of 0.8 and a specificity of 1.0 at a cut-off score of 18.44. AttnKare-D scores showed a high correlation with K-ARS scores rated by parents and experts, although the correlation was relatively low for inattention scores.Conclusion: Results of this study suggest that AttnKare-D can be a useful tool for diagnosing ADHD in children. This approach has potential to overcome limitations of current diagnostic methods, enhancing the accuracy and objectivity of ADHD diagnoses. This study lays the groundwork for further improvement and research on diagnostic tools integrating VR and AI technologies. For future clinical applications, it is necessary to conduct clinical trials involving a sufficient number of participants to ensure reliable use.

Symptoms of inattention (IA), hyperactivity, and impulsivity present in school‐aged children are important indications of developmental problems. The assessment of these symptoms largely relies on questionnaires completed by parents, teachers, and the children themselves. However, inherent perceptual biases may lead to inaccuracies in these evaluations. This study investigates the biases among different respondents for 41 elementary‐aged students (19 male and 22 female) by analyzing their responses on the attention‐deficit/hyperactivity disorder‐rating scale (ADHD‐RS) through a novel virtual reality‐based objective measure (AttnKare‐D). Correlation analysis reveals higher agreement between parents and students compared to other informant pairs. Nevertheless, when compared with the objective measure, parents and students do not exhibit distinct separation between high and low‐scoring samples, whereas teachers demonstrate significant separation on various objective measure factors. This discrepancy suggests that teachers may possess a more accurate baseline for peer‐related measurements, leading to a more objective evaluation of ADHD symptoms. However, it is essential to acknowledge that teachers are not entirely exempt from biases. Gender‐specific comparisons reveal notable differences, with male children showing the largest separation in distractibility and female children in movement. In conclusion, this study emphasizes the significance of recognizing and addressing potential biases when interpreting questionnaires assessing symptoms of IA, hyperactivity, and impulsivity in children.

Federated Learning has emerged to cope with raising concerns about privacy breaches in using Machine or Deep Learning models. This new paradigm allows the leverage of deep learning models in a distributed manner, enhancing privacy preservation. However, the server's blindness to local datasets introduces its vulnerability to model poisoning attacks and data heterogeneity, tampering with the global model performance. Numerous works have proposed robust aggregation algorithms and defensive mechanisms, but the approaches are orthogonal to individual attacks or issues. FedCC, the proposed method, provides robust aggregation by comparing the Centered Kernel Alignment of Penultimate Layers Representations. The experiment results on FedCC demonstrate that it mitigates untargeted and targeted model poisoning or backdoor attacks while also being effective in non-Independently and Identically Distributed data environments. By applying FedCC against untargeted attacks, global model accuracy is recovered the most. Against targeted backdoor attacks, FedCC nullified attack confidence while preserving the test accuracy. Most of the experiment results outstand the baseline methods.

Federated Learning is a widely adopted method for training neural networks over distributed data. One main limitation is the performance degradation that occurs when data are heterogeneously distributed. While many studies have attempted to address this problem, a more recent understanding of neural networks provides insight to an alternative approach. In this study, we show that only certain important layers in a neural network require regularization for effective training. We additionally verify that Centered Kernel Alignment (CKA) most accurately calculates similarities between layers of neural networks trained on different data. By applying CKA-based regularization to important layers during training, we significantly improved performances in heterogeneous settings. We present FedCKA, a simple framework that outperforms previous state-of-the-art methods on various deep learning tasks while also improving efficiency and scalability.

While machine learning techniques are being applied to various fields for their exceptional ability to find complex relations in large datasets, the strengthening of regulations on data ownership and privacy is causing increasing difficulty in its application to medical data. In light of this, Federated Learning has recently been proposed as a solution to train on private data without breach of confidentiality. This conservation of privacy is particularly appealing in the field of healthcare, where patient data is highly confidential. However, many studies have shown that its assumption of Independent and Identically Distributed data is unrealistic for medical data. In this paper, we propose Personalized Federated Cluster Models, a hierarchical clustering-based FL process, to predict Major Depressive Disorder severity from Heart Rate Variability. By allowing clients to receive more personalized model, we address problems caused by non-IID data, showing an accuracy increase in severity prediction. This increase in performance may be sufficient to use Personalized Federated Cluster Models in many existing Federated Learning scenarios.

Purpose: The current golden standard for attention deficit hyperactivity disorder (ADHD) diagnosis is clinical diagnosis based on psychiatric interviews and psychological examinations. This is suboptimal, as clinicians are unable to view potential patients in multiple natural settings – a necessary condition for objective diagnosis. The purpose of this paper is to improve the objective diagnosis of ADHD by analyzing a quantified representation of the actions of potential patients in multiple natural environments. Design/methodology/approach: The authors use both virtual reality (VR) and artificial intelligence (AI) to create an objective ADHD diagnostic test. Diagnostic and statistical manual of mental disorders, 5th Edition (DSM-5) and ADHD Rating Scale are used to create a rule-based system of quantifiable VR-observable actions. As a potential patient completes tasks within multiple VR scenes, certain actions trigger an increase in the severity measure of the corresponding ADHD symptom. The resulting severity measures are input to an AI model, which classifies the potential patient as having ADHD in the form inattention, hyperactivity-impulsivity, combined or neither. Findings: The result of this study shows that VR-observed actions can be extracted as quantified data, and classification of this quantified data achieves near-perfect sensitivity and specificity with a 98.3% accuracy rate on a convolutional neural network model. Originality/value: To the best of the authors’ knowledge, this is the first study to incorporate VR and AI into an objective DSM-5-based ADHD diagnostic test. By including stimulation to the visual, auditory and equilibrium senses and tracking movement and recording voice, we present a method to further the research of objective ADHD diagnosis.