Scholarlink

<p>Parkinson’s disease (PD) is a common neurodegenerative disease seen after Alzheimer’s and is shown as a growing global health issue. PD affects mainly people at a later age, and it has been seen that the impacts are more on female as compared to males. The development of PD in a person is due to the degeneration of dopamine-producing neurons in the substantia nigra part of mid brain. The reason behind the degeneration of these neurons is still unknown. The initial symptoms of PD are bradykinesia (slowed motor execution), changes in speech and handwriting, tremors, reduced automatic movements, and involuntary muscle contractions. The detection and diagnosis of PD at an early stage is very challenging, as initial symptoms are mild, so multiple clinic visits, imaging scans, and detailed neurological examinations are required. Recent innovative machine learning (ML) and deep learning (DL) techniques have made PD detection early and accurate. By employing these advanced techniques for detecting changes in speech, handwriting, EEG, and EMG signals, MRI and PET scans could be analyzed. This manuscript provides an in-depth review of research papers published from 2013 to 2015, obtained from esteemed digital libraries including IEEE, Springer, Elsevier, Hindawi, Taylor &amp; Francis, MDPI, Nature, and EAI. The thorough analysis of research papers based on preprocessing techniques, feature engineering, and classification methods was done. The manuscript examines key performance metrics, validation strategies, and widely used publicly accessible datasets. Finally, the common problems or challenges encountered by researchers and possible viable solutions are outlined.</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

Advances in artificial intelligence for Parkinson’s disease detection: a decade of machine learning and deep learning insights

  • 期刊论文

  • 发表日期：

    2026年6月15日

查看全文
  • Rohit Lamba,
  • Pooja Rani,
  • Ravi Kumar Sachdeva,
  • Priyanka Bathla

摘要

Parkinson’s disease (PD) is a common neurodegenerative disease seen after Alzheimer’s and is shown as a growing global health issue. PD affects mainly people at a later age, and it has been seen that the impacts are more on female as compared to males. The development of PD in a person is due to the degeneration of dopamine-producing neurons in the substantia nigra part of mid brain. The reason behind the degeneration of these neurons is still unknown. The initial symptoms of PD are bradykinesia (slowed motor execution), changes in speech and handwriting, tremors, reduced automatic movements, and involuntary muscle contractions. The detection and diagnosis of PD at an early stage is very challenging, as initial symptoms are mild, so multiple clinic visits, imaging scans, and detailed neurological examinations are required. Recent innovative machine learning (ML) and deep learning (DL) techniques have made PD detection early and accurate. By employing these advanced techniques for detecting changes in speech, handwriting, EEG, and EMG signals, MRI and PET scans could be analyzed. This manuscript provides an in-depth review of research papers published from 2013 to 2015, obtained from esteemed digital libraries including IEEE, Springer, Elsevier, Hindawi, Taylor & Francis, MDPI, Nature, and EAI. The thorough analysis of research papers based on preprocessing techniques, feature engineering, and classification methods was done. The manuscript examines key performance metrics, validation strategies, and widely used publicly accessible datasets. Finally, the common problems or challenges encountered by researchers and possible viable solutions are outlined.