EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multimodal Neuroimaging‐Based Machine Learning Models Leveraging Cerebral Morphometry and Glymphatic Parameters Predict Short‐Term Post‐Programming STN ‐ DBS Motor Response in Parkinson's Disease

Multimodale, auf Neuroimaging basierende Machine-Learning-Modelle mit zerebraler Morphometrie und glymphatischen Parametern zur Vorhersage des kurzfristigen post-Programming-Motoransprechens nach STN-DBS bei Parkinson-Krankheit

Yining Wang iD, Xin Zhang iD, Wenwen Yu iD, Yiyao Yang iD, Jiajun Cai, Juanjuan He, Huie Miao, Lingjuan Li, Liqin Lang, Jie Hu, Zengxin Qi iD, Liang Chen iD

CNS Neuroscience & Therapeutics · 2026-08 · 32 (8)

DOI: 10.1002/cns.71043 ↗

Auf einen Blick

Bei 155 Patient:innen mit Parkinson-Krankheit sagte ein multimodales Modell das kurzfristige Motoransprechen nach STN-DBS gut vorher. Für Tremor und das explorative Gang-Phänotyping lagen die AUCs ebenfalls hoch.

Fragestellung

Können präoperative Bildgebungs- und klinische Merkmale das kurzfristige Motoransprechen nach post-Programming STN-DBS bei Parkinson-Krankheit vorhersagen?

Methode

Retrospektive Studie mit 155 Patient:innen und 43 Kontrollen; präoperative MRT und Diffusionsbildgebung; Modellierung mit fünfgeteiltem Cross-Validation-Ansatz.

Ergebnisse

Bestes trimodales Modell: AUC 0,850 ± 0,045 (95 %-KI 0,794–0,906) für das totale Motoransprechen, 0,861 ± 0,047 (95 %-KI 0,803–0,919) für Tremor, 0,970 ± 0,019 (95 %-KI 0,946–0,994) für das explorative Gang-Phänotyping.

Einschränkungen

Externe Validierung und längere Nachbeobachtung fehlen.

Schlussfolgerung der Autor:innen

Präoperative zerebrale Morphometrie, ergänzt durch ausgewählte glymphatische Marker und klinische Basisvariablen, könnte helfen, das kurzfristige post-Programming STN-DBS-Ansprechen zu stratifizieren.

Kernaussagen

  • 155 Patient:innen mit bilateraler STN-DBS und 43 gesunde Kontrollen
  • Trimodales Modell: AUC 0,850 für das totale Motoransprechen
  • Für Gang-Phänotyping lag die AUC bei 0,970

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift CNS Neuroscience & Therapeutics, online erschienen am 22.08.2026. Verfasst von Yining Wang und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Neurosurgery of Huashan Hospital, State Key Laboratory of Medical Neurobiology, MOE Frontiers Center for Brain Science and Institutes o…. Hauptthemen laut MEDLINE: Parkinson Disease, Machine Learning, Deep Brain Stimulation, Glymphatic System. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Major Science and Technology Projects of China, National Key Research and Development Program of China, National Natural Science Foundation of China.

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background and Objectives Deep brain stimulation of the subthalamic nucleus (STN‐DBS) is effective for medication‐refractory Parkinson's disease (PD) motor symptoms, but clinical response varies across symptom domains, particularly tremor and gait. Accurate preoperative stratification is clinically important, especially for early post‐programming outcomes.

Methods

We retrospectively enrolled 155 patients with PD undergoing bilateral STN‐DBS and 43 healthy controls. Preoperative structural magnetic resonance imaging and diffusion‐weighted imaging were used to quantify brain morphometry and glymphatic markers, including diffusion tensor imaging along the perivascular space (DTI‐ALPS) and choroid plexus volume (CPV). Total motor response was evaluated in all 155 patients, tremor response in 133 patients with complete tremor subscores, and an exploratory data‐driven gait‐improvement phenotype in 66 patients with paired instrumented gait assessments. Machine‐learning models were developed using fold‐wise feature selection and hyperparameter tuning and were evaluated by fivefold cross‐validation.

Results

Best‐performing trimodal models yielded AUCs of 0.850 ± 0.045 (95% CI, 0.794–0.906) for total motor response, 0.861 ± 0.047 (95% CI, 0.803–0.919) for tremor response, and 0.970 ± 0.019 (95% CI, 0.946–0.994) for the exploratory gait‐improvement phenotype. Morphometric‐only models retained substantial predictive performance, with maximum AUCs of 0.830, 0.849, and 0.955 for the motor, tremor, and gait‐related endpoints, respectively. For the exploratory gait phenotype, clinical‐plus‐morphometric and trimodal models performed similarly, suggesting limited incremental value of glymphatic variables in this subgroup.

Conclusion

Preoperative cerebral morphometry, complemented by selected glymphatic markers and baseline clinical variables, may help stratify short‐term post‐programming STN‐DBS response in PD. These findings support further development of imaging‐informed DBS outcome prediction, while external validation and longer‐term follow‐up remain necessary before clinical implementation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Parkinson DiseaseMachine LearningDeep Brain StimulationGlymphatic SystemSubthalamic NucleusNeuroimagingHumansFemaleMaleMiddle AgedRetrospective StudiesAgedPredictive Learning ModelsMultimodal ImagingDiffusion Tensor ImagingTreatment OutcomeMagnetic Resonance Imaging

Stichwörter der Autor:innen

Parkinson diseaseSTN‐DBSglymphatic parametersmachine learningmorphometrics

Autor:innen und Institutionen

  • Department of Neurosurgery of Huashan Hospital, State Key Laboratory of Medical Neurobiology, MOE Frontiers Center for Brain Science and Institutes of Brain Science, Fudan University, Shanghai, China.
  • Shanghai Key Laboratory of Brain Function and Restoration and Neural Regeneration, Shanghai, China.
  • Shanghai Clinical Medical Center of Neurosurgery, Shanghai, China.
  • National Center for Neurological Disorders, Shanghai, China.
  • Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University, Shanghai, China.
  • Department of Nursing, Huashan Hospital, Fudan University, Shanghai, China.

ORCID-Profile: Yining Wang, Xin Zhang, Wenwen Yu, Yiyao Yang, Zengxin Qi, Liang Chen

Förderung

  • National Major Science and Technology Projects of China
  • National Key Research and Development Program of China
  • National Natural Science Foundation of China
  • Natural Science Foundation of Shanghai Municipality
  • Natural Science Support Program of XPCC
  • Special Program for Scientific and Technological Innovation of 3rd Division General Hospital of XPCC
  • Science and Technology Commission of Shanghai municipality frontier innovation program
  • Shanghai Municipal Science and Technology Major Project

Zitieren

APA

Wang, Y., Zhang, X., Yu, W., Yang, Y., Cai, J., He, J., Miao, H., Li, L., Lang, L., Hu, J., Qi, Z., & Chen, L. (2026). Multimodal Neuroimaging‐Based Machine Learning Models Leveraging Cerebral Morphometry and Glymphatic Parameters Predict Short‐Term Post‐Programming STN ‐ DBS Motor Response in Parkinson's Disease. CNS Neuroscience & Therapeutics, 32(8). https://doi.org/10.1002/cns.71043

Vancouver

Wang Y, Zhang X, Yu W, Yang Y, Cai J, He J, et al. Multimodal Neuroimaging‐Based Machine Learning Models Leveraging Cerebral Morphometry and Glymphatic Parameters Predict Short‐Term Post‐Programming STN ‐ DBS Motor Response in Parkinson's Disease. CNS Neuroscience & Therapeutics. 2026;32(8). doi:10.1002/cns.71043

BibTeX
@article{Wang2026,
  author = {Wang, Yining and Zhang, Xin and Yu, Wenwen and Yang, Yiyao and Cai, Jiajun and He, Juanjuan and Miao, Huie and Li, Lingjuan and Lang, Liqin and Hu, Jie and Qi, Zengxin and Chen, Liang},
  title = {Multimodal Neuroimaging‐Based Machine Learning Models Leveraging Cerebral Morphometry and Glymphatic Parameters Predict Short‐Term Post‐Programming STN ‐ DBS Motor Response in Parkinson's Disease},
  journal = {CNS Neuroscience & Therapeutics},
  year = {2026},
  volume = {32},
  number = {8},
  doi = {10.1002/cns.71043}
}
Kostenlos starten →