Scholarlink

<p>Parkinson’s disease (PD) affects both motor and non-motor functions, often progressing gradually and remaining undiagnosed in its early stages. To support research aimed at earlier and more accessible detection, we introduce UT-MoDaPark, a multimodal dataset collected through a mobile application designed to capture a broad spectrum of PD-related symptoms. The dataset includes recordings from 151 participants (68 individuals with clinically confirmed PD and 83 healthy controls) collected in structured, in-person sessions to ensure consistency and reliability. Participants engaged in five interactive modules using a tablet: a drawing task for assessing fine motor control, a selfie-based facial expression task, a voice-based emotion recognition task, a structured symptom questionnaire, and a gamified voice control exercise. The released dataset includes drawing trajectories, facial landmarks and Action Units, voice-derived features, questionnaire responses, and participant metadata. UT-MoDaPark offers a unified and symptom-rich resource that can support the development of digital tools for early detection, remote monitoring, and personalized assessment of PD, with particular relevance to real-world, at-home, or low-resource healthcare settings.</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 multimodal mobile dataset for Parkinson’s disease symptom assessment

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月22日

查看全文
  • Atefeh Irani,
  • Yasaman Haghbin,
  • Maryam Dadkhah,
  • Farnaz Sedaghati,
  • Saeed Maroof,
  • Mohammad Rohani,
  • Maziar Emamikhah,
  • Seyed Amir Hassan Habibi,
  • Reshad Hosseini,
  • Hadi Moradi

摘要

Parkinson’s disease (PD) affects both motor and non-motor functions, often progressing gradually and remaining undiagnosed in its early stages. To support research aimed at earlier and more accessible detection, we introduce UT-MoDaPark, a multimodal dataset collected through a mobile application designed to capture a broad spectrum of PD-related symptoms. The dataset includes recordings from 151 participants (68 individuals with clinically confirmed PD and 83 healthy controls) collected in structured, in-person sessions to ensure consistency and reliability. Participants engaged in five interactive modules using a tablet: a drawing task for assessing fine motor control, a selfie-based facial expression task, a voice-based emotion recognition task, a structured symptom questionnaire, and a gamified voice control exercise. The released dataset includes drawing trajectories, facial landmarks and Action Units, voice-derived features, questionnaire responses, and participant metadata. UT-MoDaPark offers a unified and symptom-rich resource that can support the development of digital tools for early detection, remote monitoring, and personalized assessment of PD, with particular relevance to real-world, at-home, or low-resource healthcare settings.