Closed-Loop Deep Brain Stimulation — PatSnap Eureka
Closed-Loop Deep Brain Stimulation Pipeline in Parkinson's Disease & Dystonia
Adaptive DBS is transitioning from bench to bedside: sensing-capable pulse generators now enable real-world validation of biomarker-driven feedback algorithms for movement disorders. Explore the full innovation landscape with PatSnap Eureka.
Pathological Synchronization as the Therapeutic Target
Parkinson's disease motor pathophysiology centres on excessive oscillatory power in the basal ganglia-thalamo-cortical circuit, particularly beta-band (13–35 Hz) local field potentials (LFPs) recorded from the subthalamic nucleus (STN) and globus pallidus internus (GPi). The STN-GPe excitatory-inhibitory loop is consistently identified as the primary oscillatory generator underlying bradykinesia, rigidity, tremor, and freezing of gait (FoG). Foundational work on beta oscillations as a therapeutic target is tracked by organisations including NIH's National Institute of Neurological Disorders and Stroke and WHO, which recognises PD among the fastest-growing neurological disorders globally.
University of Oxford's Nuffield Department of Clinical Neurosciences identifies beta-frequency LFP power as correlating with bradykinesia and rigidity, and as a viable control variable for adaptive DBS. A University of Minnesota study demonstrates that pallidal beta power (13–30 Hz) correlates directly with bradykinesia severity, and that stimulation can suppress or amplify specific frequency-selective (16–22 Hz) neural activity in real-time in human subjects. The PatSnap Analytics platform provides comprehensive patent landscape analysis across neuromodulation IP.
Beyond the STN and GPi, the pedunculopontine nucleus (PPN) emerges as a candidate target for medication-refractory FoG, with clinical trial evidence from the University of Florida (NCT02318927). Dystonia is addressed as a secondary indication: GPi is named as the primary DBS target for both PD and dystonia, with STN-DBS models including dystonia in quantitative modelling frameworks from the University of Sydney. Neurochemical biomarkers — dopamine, serotonin, and adenosine — are highlighted by Mayo Clinic as detectable via electrochemical microsensors for closed-loop DBS control, opening a distinct modality beyond electrophysiological LFPs. Learn how PatSnap supports life sciences R&D teams navigating complex drug-device pipelines.
Eight Distinct Closed-Loop DBS Approaches Across the Pipeline
From commercially validated LFP-guided adaptive DBS to neurochemical feedback and spinal cord stimulation — the modality landscape spans proof-of-concept through early chronic clinical use.
Adaptive / Closed-Loop DBS Using LFP Biomarkers
Stimulation amplitude or timing is modulated in real time based on beta-band LFP power at the STN or GPi. A University College London proof-of-concept in 8 PD patients demonstrated improved efficacy and efficiency versus continuous DBS. The Medtronic Percept™ PC enables chronic condition-dependent stimulation in implanted patients. The AlphaDBSR system (NCT04681534) demonstrated artifact-free simultaneous sensing and stimulation in 3 PD patients.
Early chronic clinical usePhase-Responsive & Phasic Burst Stimulation (PhaBS / eiDBS)
Rather than titrating amplitude to oscillation power, bursts of pulses are delivered at specific phases of the pathological oscillation, optimised using patient-specific phase response curves (PRCs). University of Minnesota's Evoked Interference DBS (eiDBS) demonstrates real-time causal control of 16–22 Hz GPi activity in a human PD patient.
First-in-human proof of conceptDelayed Feedback Desynchronizing Stimulation
Jülich Research Center computationally derived protocols combine amplitude-modulated high-frequency pulse trains with nonlinear delayed feedback to desynchronize STN-GPe networks. Stanford University extended this to multisite delayed feedback applied across multiple brain sites to restore desynchronized neuronal activity.
Computational / PreclinicalNeurofeedback-Enabled Volitional Beta Control
PD patients learn to volitionally reduce STN beta oscillations via real-time visual neurofeedback from implanted DBS electrodes, with associated improvements in hand movement speed. University of Zurich demonstrated control persisting after removal of visual feedback. A fully implanted Medtronic Percept system study in 8 PD patients extended this to bidirectional beta modulation.
Early clinical feasibilityMachine Learning & Neural Network-Based Control
Classical PID controllers are replaced with backpropagation neural networks (BPNN), Bayesian adaptive dual controllers, active learning frameworks, and gradient-boosted tree models. A Georgia Tech/Emory active learning framework minimises experiments needed to identify DBS parameter-neural response links. Heinrich Heine University gradient-boosted models achieve r > 0.8 correlation with DBS outcome using only five coherence features.
Computational / SimulationNeurochemical Closed-Loop DBS
Electrochemical biosensors detect neurotransmitter fluctuations — dopamine, serotonin, and adenosine — as the feedback signal for DBS control, rather than electrophysiological LFPs. Mayo Clinic proposes stimulation-evoked neurotransmitter changes measurable in real time as individualised biomarkers tailoring DBS therapy to disease progression and electrode-tissue interface changes.
Preclinical / Early translationalKey Quantitative Signals from the Closed-Loop DBS Literature
Clinical cohort sizes, biomarker performance metrics, and development stage distribution — all derived from patent and literature analysis via PatSnap Eureka.
Clinical Evidence: Patient Cohort Sizes by Study
Charité StimFit (35 patients) and Heinrich Heine MEG+LFP ML (36 patients) represent the largest cohorts; early-stage closed-loop trials range from 3–8 patients.
Modality Distribution by Development Stage
Most closed-loop DBS modalities remain at computational or preclinical stage; adaptive LFP-driven aDBS and neurofeedback have reached early clinical use.
Eight Key Clinical Evidence Milestones in Closed-Loop DBS
From first-in-human feasibility through randomised crossover trials — the clinical evidence base for adaptive DBS is expanding rapidly across multiple institutions.
| Study / Trial | Institution | N | Key Outcome | Stage |
|---|---|---|---|---|
| UCL aDBS BCI (2013) | University College London | 8 | aDBS superior efficacy and efficiency vs continuous DBS; reduced stimulation time | PoC |
| Würzburg 8-hour aDBS (2018) | University of Würzburg | 8 | UPDRS 30.5 ± 3.4 vs 22.2 ± 3.3 (p = 0.003); beta-LFP r = 0.506 with clinical state | PoC |
| Juntendo Percept™ PC Chronic aDBS (2021) | Juntendo University | 1 | Chronic condition-dependent stimulation without stimulation-induced side effects | Early Clinical |
| NCT04681534 AlphaDBSR (2021) | Maastricht University | 3 | First-in-human artifact-free simultaneous sensing and closed-loop stimulation | PoC |
| NCT02318927 PPN CL-DBS (2021) | University of Florida | 5 | Primary outcome: ≥40% FoG improvement in ≥60% of subjects at 6 months | Early Clinical |
| TOPS® Randomized Study (2022) | Cleveland Clinic | 26 | Comparable motor improvement and side-effect profiles; TOPS well-tolerated | Clinical |
Track every clinical trial signal in adaptive DBS
PatSnap Eureka monitors NCT registrations, publications, and patent filings across the full neuromodulation pipeline.
Convergent Innovation: Multi-Target, Drug-Device, and Multimodal Strategies
Retrieved results signal several convergent combination strategies that go beyond single-target, single-modality closed-loop DBS.
Multi-Target Closed-Loop DBS (STN + PPN; STN + SNr)
Beijing Tiantan Hospital describes bilateral STN + PPN combined DBS for overlapping FoG and bradykinesia. A 2023 computational study evaluates concurrent STN + substantia nigra pars reticulata (SNr) DBS for FoG via cortico-subcortical network modelling. University of Oxford develops mathematical frameworks for multiple independently controlled DBS contacts, enabling precision targeting of multiple pathological populations. Explore PatSnap life sciences solutions for multi-target IP analysis.
Drug-Device Combination: Closed-Loop DBS + RL Medication Optimisation
A University of Texas computational study demonstrates that reinforcement learning (RL) can derive optimal sequential medication regimens — levodopa, dopamine agonist, and other PD drugs — across 431 PD patients over 55.5-month follow-up, using Markov decision process optimisation. This signals future integration with closed-loop DBS to co-optimise pharmacological and electrophysiological therapy in real time. The combined DBS + L-DOPA approach is documented as the most effective current standard, with DBS enabling levodopa dose reduction and delaying dyskinesia onset.
Research-Driven Innovation: Academic Centres Lead the Pipeline
Innovation activity in this dataset is predominantly literature-driven (academic papers) rather than patent-driven, reflecting the research-heavy, pre-commercialisation stage of most closed-loop DBS algorithms. No patents were identified in the retrieved results. The PatSnap Analytics platform provides the tools to monitor when academic innovation converts to IP filings across neuromodulation.
University of Oxford / MRC Brain Network Dynamics Unit contributes multiple high-impact papers on beta biomarkers, aDBS proof-of-concept in patients, and the translational roadmap for aDBS. University of Minnesota produces concentrated output on phase-responsive DBS, Bayesian adaptive control, Phasic Burst Stimulation, eiDBS human proof of concept, and patient-specific pathway activation models. University of Florida (Norman Fixel Institute) contributes clinical trial evidence for closed-loop PPN-DBS in FoG and broad DBS hardware and software reviews.
Engineering and computational groups — including Tianjin University (BPNN controllers), Zhejiang University (robust adaptive state-space control), and Georgia Tech/Emory (active learning) — are developing the algorithmic foundations for next-generation systems. Hospital-based clinical groups at Juntendo University (first chronic aDBS with Percept™ PC) and Maastricht University (AlphaDBSR first-in-human) are generating the earliest real-world evidence. The PatSnap customer success stories show how R&D teams use Eureka to track academic-to-commercial IP transitions. Regulatory guidance on neuromodulation devices is available from the FDA.
Closed-Loop Deep Brain Stimulation — Key Questions Answered
Adaptive DBS (aDBS) modulates stimulation amplitude or timing in real time based on local field potential (LFP) power, typically beta-band activity recorded at the STN or GPi. Evidence spans proof-of-concept through early chronic clinical use, including commercially implanted sensing-capable pulse generators such as the Medtronic Percept PC.
Beta-band LFP power (13–35 Hz) recorded from the STN or GPi is the most consistently validated feedback signal across retrieved results. However, limitations exist: beta-driven closed-loop DBS can interfere with volitional movements, as beta naturally desynchronizes during movement initiation.
Key clinical signals include NCT02318927 (University of Florida, PPN closed-loop DBS for freezing of gait, 5 patients), NCT04681534 (Maastricht University, AlphaDBSR first-in-human, 3 patients), the University of Würzburg 8-hour aDBS study (8 patients, UPDRS 30.5 vs 22.2, p=0.003), and the Charité StimFit randomized crossover trial (35 PD patients).
Multiple retrieved results describe replacing classical PID controllers with data-driven algorithms including backpropagation neural networks (BPNN), Bayesian adaptive dual controllers, active learning frameworks, and gradient-boosted tree models. Heinrich Heine University demonstrated that gradient-boosted tree models based on subthalamo-cortical coherence features achieve r > 0.8 correlation with DBS outcome using only five coherence features.
The pedunculopontine nucleus (PPN) emerges as a specific target for medication-refractory freezing of gait (FoG). Retrieved results from both Beijing Tiantan Hospital (bilateral STN + PPN DBS case) and University of Florida (NCT02318927 clinical trial) describe mixed but promising outcomes. A Shanghai University of Electric Power neural mass model finds PPN-LFP-guided STN stimulation effective at suppressing pathological oscillations.
Yes. Multiple retrieved results address DBS as a complement to levodopa and dopamine agonists. The combination is described as the current evidence-based standard, with closed-loop DBS enabling reduced levodopa equivalent daily dosage (LEDD) in some reports. A University of Texas computational study also demonstrates that reinforcement learning can derive optimal sequential medication regimens, suggesting future integration with closed-loop DBS to co-optimize pharmacological and electrophysiological therapy in real time.
Still have questions? Let PatSnap Eureka search the patent and literature database for you.
Ask PatSnap Eureka About Closed-Loop DBSAccelerate Your Adaptive DBS Innovation Intelligence
Join 18,000+ innovators already using PatSnap Eureka to navigate complex drug-device pipelines, track clinical signals, and identify white-space opportunities in neuromodulation.
References
- University of Oxford — What brain signals are suitable for feedback control of deep brain stimulation in Parkinson's disease? (2012)
- University of Minnesota — Controlling pallidal oscillations in real-time in Parkinson's disease using evoked interference deep brain stimulation (eiDBS) (2022)
- Zhejiang University — Robust Adaptive Deep Brain Stimulation Control of Non-Stationary Cortex-Basal Ganglia-Thalamus Network Models in Parkinson's Disease (2023)
- University of Florida — Closed-Loop Deep Brain Stimulation to Treat Medication-Refractory Freezing of Gait in Parkinson's Disease (2021)
- Mayo Clinic — A neurochemical closed-loop controller for deep brain stimulation: toward individualized smart neuromodulation therapies (2014)
- Heinrich Heine University Düsseldorf — Neuronal oscillations predict deep brain stimulation outcome in Parkinson's disease (2022)
- University College London — Adaptive deep brain stimulation in advanced Parkinson disease (2013)
- University of Würzburg — Eight-hours adaptive deep brain stimulation in patients with Parkinson disease (2018)
- Juntendo University — Case Report: Chronic Adaptive Deep Brain Stimulation Personalizing Therapy Based on Parkinsonian State (2021)
- Maastricht University — A New Implantable Closed-Loop Clinical Neural Interface: First Application in Parkinson's Disease (2021)
- University of Minnesota — Phasic Burst Stimulation: A Closed-Loop Approach to Tuning Deep Brain Stimulation Parameters for Parkinson's Disease (2016)
- Jülich Research Center — Pulsatile desynchronizing delayed feedback for closed-loop deep brain stimulation (2017)
- Stanford University — Multisite Delayed Feedback for Electrical Brain Stimulation (2018)
- University of Zurich — Deep brain electrical neurofeedback allows Parkinson patients to control pathological oscillations and quicken movements (2020)
- Tianjin University — Neural Network-Based Closed-Loop Deep Brain Stimulation for Modulation of Pathological Oscillation in Parkinson's Disease (2020)
- University of Minnesota — Bayesian adaptive dual control of deep brain stimulation in a computational model of Parkinson's disease (2018)
- Georgia Tech / Emory University — An active learning framework for personalized deep brain stimulation (2023)
- Duke University — Closed-loop spinal cord stimulation is superior in restoring locomotion in rodent models of Parkinson's Disease (2022)
- University of Texas Health Science Center at Houston — Computational medication regimen for Parkinson's disease using reinforcement learning (2021)
- Bern University Hospital — Combining Multimodal Biomarkers to Guide Deep Brain Stimulation Programming in Parkinson Disease (2023)
- EPFL / Lausanne University Hospital — Beta-driven closed-loop deep brain stimulation can compromise human motor behavior in Parkinson's Disease (2019)
- Royal Melbourne / Austin Hospitals — Tailoring Subthalamic Nucleus Deep Brain Stimulation for Parkinson's Disease Using Evoked Resonant Neural Activity (2020)
- Cleveland Clinic — Temporally optimized patterned stimulation (TOPS®) as a therapy to personalize deep brain stimulation treatment of Parkinson's disease (2022)
- Charité Berlin — Automated deep brain stimulation programming based on electrode location – a randomized, cross-over trial (2023)
- NIH National Institute of Neurological Disorders and Stroke — Parkinson's Disease Research
- World Health Organization — Neurological Disorders: Public Health Challenges
- U.S. Food and Drug Administration — Neuromodulation Device Regulatory Guidance
All data and statistics on this page are sourced from the references above and from PatSnap's proprietary innovation intelligence platform. This report is derived from a limited set of patent and literature records retrieved across targeted searches and represents a snapshot of innovation signals within this dataset only.
PatSnap Eureka searches patents and research to answer instantly.