EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Framework for the Prediction of Parkinson’s Disease Using Agentic Artificial Intelligence

Areej fatima, Rabia Tehseen, Kashif Nasr, Uzma Omer, Muhammad Inam Ul Haq, Rubab Javaid

International Journal of Innovations in Science and Technology · 02.12.2025 · 3033

DOI: 10.33411/ijist/1665 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Innovations in Science and Technology, online erschienen am 02.12.2025. Verfasst von Areej fatima und 5 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Parkinson’s disease (PD) is a progressive neurodegenerative disorder that is difficult to diagnose, particularly in its early stages. Subtle, slowly evolving symptoms often delay confirmation, reducing opportunities for timely intervention that could improve outcomes and quality of life. Conventional diagnosis relies largely on clinical observation, which can be subjective and insufficiently sensitive for early detection. This thesis proposes an Agentic Artificial Intelligence (AAI) framework for early PD detection and severity assessment using voice-based biomarkers. Biomedical voice parameters are leveraged because vocal changes can reflect early neurological impairment. Two publicly available Kaggle datasets containing voice recordings from individuals with PD and healthy controls are used to train and evaluate the models. For detection, an XGBoost classifier achieves 94.68% accuracy with strong discriminative performance. For severity estimation, XGBoost regression models predict motor and total Unified Parkinson’s Disease Rating Scale (UPDRS) scores with high agreement to clinically reported measurements. A key contribution is an agentic decision-making layer that autonomously interprets model outputs, performs disease staging, and generates stage-dependent monitoring and treatment recommendations. Unlike conventional predictive pipelines that stop at numerical outputs, the proposed system translates predictions into actionable clinical insights to support structured decision-making. Experimental results indicate that the framework can detect PD and estimate severity effectively from non-invasive voice data, highlighting the potential of AAI for earlier diagnosis, personalized monitoring, and intelligent clinical decision support in healthcare. The multi-layer design supports modular updates to models and agent policies, enabling telehealth deployment and longitudinal tracking as additional voice samples become available over time

Zitieren

APA

fatima, A., Tehseen, R., Nasr, K., Omer, U., Haq, M. I. U., & Javaid, R. (2025). A Framework for the Prediction of Parkinson’s Disease Using Agentic Artificial Intelligence. International Journal of Innovations in Science and Technology, 3033. https://doi.org/10.33411/ijist/1665

Vancouver

fatima A, Tehseen R, Nasr K, Omer U, Haq MIU, Javaid R. A Framework for the Prediction of Parkinson’s Disease Using Agentic Artificial Intelligence. International Journal of Innovations in Science and Technology. 2025:3033. doi:10.33411/ijist/1665

BibTeX
@article{fatima2025,
  author = {fatima, Areej and Tehseen, Rabia and Nasr, Kashif and Omer, Uzma and Haq, Muhammad Inam Ul and Javaid, Rubab},
  title = {A Framework for the Prediction of Parkinson’s Disease Using Agentic Artificial Intelligence},
  journal = {International Journal of Innovations in Science and Technology},
  year = {2025},
  pages = {3033},
  doi = {10.33411/ijist/1665}
}
Kostenlos starten →