Scholarlink

<p>Affective disorders, including Major Depressive Disorder (MDD) and Bipolar Disorder (BD), exhibit significant mood abnormalities, making rapid diagnosis essential for social stability and healthcare efficiency. Traditional diagnostic solutions, including medical history collection and psychological assessments, often struggle to differentiate their similar clinical presentations, leading to time-consuming, laborious, and a high rate of misdiagnosis. Here, we propose Emoface, an AI-assisted diagnostic model that reads the emotional activities of faces in affective disorders. By analyzing videos from 353 participants exposed to various emotional stimuli, Emoface identified unique facial digital biomarkers distinguishing BD from MDD. Based on this, Emoface contributed to develop the first digital facial mapping for clinical and teaching use. In clinical practice with 347 patients, Emoface achieved precise diagnosis based on various facial visual signals, with accuracy rates of 95.29% for BD and 87.05% for MDD, offering a reliable face-based AI solution in a new era of affective disorders.</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

Emoface: AI-assisted diagnostic model for differentiating major depressive disorder and bipolar disorder via facial biomarkers

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月18日

查看全文
  • Jiahui Yu,
  • Jingkai Chen,
  • Yutong Zhang,
  • Hailong Lyu,
  • Tianyu Ma,
  • Huimin Huang,
  • Zhong Wang,
  • Xin Xu,
  • Shaohua Hu,
  • Yingke Xu

摘要

Affective disorders, including Major Depressive Disorder (MDD) and Bipolar Disorder (BD), exhibit significant mood abnormalities, making rapid diagnosis essential for social stability and healthcare efficiency. Traditional diagnostic solutions, including medical history collection and psychological assessments, often struggle to differentiate their similar clinical presentations, leading to time-consuming, laborious, and a high rate of misdiagnosis. Here, we propose Emoface, an AI-assisted diagnostic model that reads the emotional activities of faces in affective disorders. By analyzing videos from 353 participants exposed to various emotional stimuli, Emoface identified unique facial digital biomarkers distinguishing BD from MDD. Based on this, Emoface contributed to develop the first digital facial mapping for clinical and teaching use. In clinical practice with 347 patients, Emoface achieved precise diagnosis based on various facial visual signals, with accuracy rates of 95.29% for BD and 87.05% for MDD, offering a reliable face-based AI solution in a new era of affective disorders.