Scholarlink

<p>Generalized Anxiety Disorder (GAD) is a common mental health illness that occurs because of constant and extreme worry or tension. For timely and effective treatment, it is very important to detect this disorder on time. Thus, the objective of this study is to detect GAD using several deep learning-based approaches such as Convolution Neural Network (CNN) and a hybrid approach consists of CNN and Residual Networks (ResNet). As compared to traditional methods, this study focuses on exploration of advance deep learning techniques for timely assessment of GAD. In order to achieve this objective, Distress Analysis Interview Corpus (DAIC) data has been used which consists of audio recordings of 189 participants along with their questionnaire responses. A comparative analysis of CNN and ResNet-like approach has been done using several performance metrics such as precision, recall, and accuracy. Results have shown that hybrid approach outperforms the standalone CNN based approach with an accuracy of 81.23%. This study has underscored the potential of ResNet-like approach in significantly assessing GAD. Future innovations can also be done to propose an advance deep learning approach for automatic detection and prevention of deterioration of GAD symptoms or other kind of anxiety disorders along with several optimization techniques.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Deep Learning-Based Technique for Detection of Generalized Anxiety Disorder using CNN and ResNet-like Approach

  • 期刊论文

  • 发表日期：

    2025年3月17日

查看全文
  • Rakhi Nagpal,
  • Saravjeet Singh,
  • Aditi Moudgil

摘要

Generalized Anxiety Disorder (GAD) is a common mental health illness that occurs because of constant and extreme worry or tension. For timely and effective treatment, it is very important to detect this disorder on time. Thus, the objective of this study is to detect GAD using several deep learning-based approaches such as Convolution Neural Network (CNN) and a hybrid approach consists of CNN and Residual Networks (ResNet). As compared to traditional methods, this study focuses on exploration of advance deep learning techniques for timely assessment of GAD. In order to achieve this objective, Distress Analysis Interview Corpus (DAIC) data has been used which consists of audio recordings of 189 participants along with their questionnaire responses. A comparative analysis of CNN and ResNet-like approach has been done using several performance metrics such as precision, recall, and accuracy. Results have shown that hybrid approach outperforms the standalone CNN based approach with an accuracy of 81.23%. This study has underscored the potential of ResNet-like approach in significantly assessing GAD. Future innovations can also be done to propose an advance deep learning approach for automatic detection and prevention of deterioration of GAD symptoms or other kind of anxiety disorders along with several optimization techniques.