.png){kind=icon}
返回
A Unified Deep Learning Framework for Smartphone-Enabled ADHD Detection
.png){kind=icon}
PRE
AI
DOI：10.1109/TIM.2023.3338713.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Attention-deficit hyperactivity disorder (ADHD) is a persistent condition with repeated issues, such as difficulty maintaining attention, impetuous behavior, and hyperactivity. It severely impacts a considerable population of children. Therefore, it is crucial to focus on early detection and treatment of ADHD. Few recent studies utilize electroencephalogram (EEG) measurements acquired during distinct cognitive graphics along with machine learning and deep learning for its detection. However, such studies are limited with the availability of only two public EEG measurement datasets to date. In view of these aspects, this work proposes a deep learning-based unified framework for the detection of ADHD and its real-time deployment on smartphones. Extensive performance assessment with the average accuracies of 99.02% and 98.60% for EEG measurements in two public datasets suggests that the proposed approach can be more accurate than those of prior studies in some cases. Robustness analysis using inter-dataset signals further demonstrates the effectiveness and cost efficiency of the proposed framework. The framework predicts ADHD/non-ADHD signals on smartphone with a latency of 67 ms and utilizes an overall energy of 82.50 mWh for processing EEG measurements, detecting ADHD/non-ADHD and communicating these details to the medical server. This demonstrates its feasibility in real-time environment with the facility of tracking ADHD progression by subjects toward self-assessment, thereby decreasing the transportation cost of frequent clinical sessions.
Keyword:
Attention-deficit hyperactivity disorder (ADHD)
convolutional neural network (CNN)
electroencephalogram (EEG)
期刊
IF:
5.9
论文数:
2.0W
被引数:
5.8W
机构
引用论文
Complex syntax acquisition: A longitudinal case study of a child with specific language impairment
PREAI
分享收藏
Myocardial Infarction Severity Stages Classification From ECG Signals Using Attentional Recurrent Neural Network.png)
IEEE SENSORS JOURNAL
IEEE SENSORS JOURNALIF4.5
PREAI
分享收藏Automated Classification of Retinal OCT Images Using a Deep Multi-Scale Fusion CNNIEEE SENSORS JOURNAL
IEEE SENSORS JOURNALIF4.5
PREAI
分享收藏
查看更多内容