Artificial Intelligence-Driven Neuromodulation in Neurodegenerative Disease: Precision in Chaos, Learning in Loss.
Artificial Intelligence-Driven Neuromodulation in Neurodegenerative Disease: Precision in Chaos, Learning in Loss.
Where did the research take place?
The study site has not been established. Author addresses may differ from where the research occurred.
Centro, IT · Author affiliation
IRCCS Centro Neurolesi Bonino-Pulejo, S.S. 113 Via Palermo, C.da Casazza, 98124 Messina, Italy.Location evidence
Palermo, IT · Author affiliation
IRCCS Centro Neurolesi Bonino-Pulejo, S.S. 113 Via Palermo, C.da Casazza, 98124 Messina, Italy.Location evidence
Messina, IT · Author affiliation
IRCCS Centro Neurolesi Bonino-Pulejo, S.S. 113 Via Palermo, C.da Casazza, 98124 Messina, Italy.Location evidence
A plain-language reading has not been prepared for this paper yet.
Original abstract
Neurodegenerative disorders such as Alzheimer's disease (AD), Parkinson's disease (PD), and multiple sclerosis (MS) are marked by progressive network dysfunction that challenges conventional, protocol-based neurorehabilitation. In parallel, neuromodulation, encompassing deep brain stimulation (DBS), transcranial magnetic stimulation (TMS), transcranial direct current stimulation (tDCS), vagus nerve stimulation (VNS), and artificial intelligence (AI), has matured rapidly, offering complementary levers to tailor therapy in real time. This narrative review synthesizes current evidence at the intersection of AI and neuromodulation in neurorehabilitation, focusing on how data-driven models can personalize stimulation and improve functional outcomes. We conducted a targeted literature synthesis of peer-reviewed studies identified via PubMed, Embase, Scopus, and reference chaining, prioritizing recent clinical and translational reports on adaptive/closed-loop systems, predictive modeling, and biomarker-guided protocols. Across indications, convergent findings show that AI can optimize device programming, enable state-dependent stimulation, and support clinician decision-making through multimodal biomarkers derived from neural, kinematic, and behavioral signals. Key barriers include data quality and interoperability, model interpretability and safety, and ethical and regulatory oversight. Here we argue that AI-enhanced neuromodulation reframes neurorehabilitation from static dosing to adaptive, patient-specific care. Advancing this paradigm will require rigorous external validation, standardized reporting of control policies and artifacts, clinician-in-the-loop governance, and privacy-preserving analytics.