AI and Machine Learning-Driven Precision Diagnosis and Treatment for Parkinson's Disease and Related Disorders: A Systematic Review of Quantitative Ultrasound Imaging, Personalized Treatment Strategies, and Adaptive Deep Brain Stimulation

Abstract

Parkinson's disease (PD) and related illnesses can now be better detected and treated because to recent developments in artificial intelligence (AI) and machine learning (ML). PD is a degenerative neurological condition that profoundly impairs both motor and non-motor function, making prompt diagnosis and effective treatment difficult. This systematic review focuses on the examination of the recent literature (2023-2026) in which the authors consider different aspects of the application of AI for detection and therapy of the condition in question through diverse data modalities such as medical imaging, speech signals, clinical biomarkers, and electrophysiology. It has been found that more sophisticated ML algorithms such as deep learning, ensemble learning, and multimodal fusion contribute to improvement in the accuracy of diagnostics. Adaptive Deep Brain Stimulation (aDBS) has also been examined as a prospective solution to provide personalized treatment by allowing controlling the parameters of stimulation in real time. At the same time, certain challenges related to limitations in datasets, interpretability and clinical validation have not been overcome yet.