Movement-related symptoms in Parkinson’s disease (PD) and other neurological conditions arise from irregular electrical signals in the brain regions responsible for motor control. Deep Brain Stimulation (DBS), approved since 1997 by the FDA, presents a transformative treatment option for PD patients as medications lose efficacy. By combining brain-implanted electrodes with a pacemaker-like device implanted under the skin of the upper chest, patients often achieve remarkable symptom relief.

Although approved for over 30 years, the current DBS therapy still lacks adaptability to fluctuating symptom severity levels in neurological disorders. This can lead to challenges for patients and their quality of life, as the constant level of stimulation doesn’t match symptom volatility. Furthermore, DBS protocols need constant adjustment over time to maintain a good therapeutic effect and avoid the side effects induced by overstimulation. Next generation adaptive DBS therapy, in which stimulation strength adjusts automatically to symptoms, promises to solve these problems but their development is severely hampered by the extreme amounts of clinician time required in tailoring such approaches to specific individuals.

Deep Brain Decode aims to solve this problem and usher in the mainstream use of adaptive DBS by using machine learning to estimate a patient’s current symptoms state and dynamically adjust electrical stimulation therapy in real-time. This digital solution is possible thanks to a database of invasive recordings of patients with various neurological disorders accrued across five years, all in one of the world’s leading deep brain stimulation research centers.

The recordings enabled the team to build decoding models and demonstrate proof-of-concept validation of generalized neural decoders that can be integrated in a plug-and-play fashion into deep brain stimulation implants. Critically, the unique algorithms developed by the team allow for training-free personalization of the stimulation, removing the by far biggest obstacle to implementing adaptive DBS – the unscalable demands on the clinician’s time required to personalize and implement adaptive techniques.

Team Deep Brain Decode is fully based at Charité and consists of a neuroscientist and expert in machine learning, a medical doctor experienced in conducting neurophysiology recordings and data analysis, and a key expert in the field of DBS, movement disorders, and neurotechnology.

Deep Brain Decode aims to enable next generation therapy for neurological disorders by providing medical device companies with generalized neural decoding models.