Diabetes Dialogue: Non-Invasive EEG Sensor Aims to Track Glucose in Real Time

Welcome back to Diabetes Dialogue: Technology, Therapeutics, & Real-World Perspectives!

Non-invasive glucose monitoring is entering a new phase, with an EEG-based wearable positioned to sense blood sugar through brain rhythms rather than interstitial fluid. Early data from the developer suggest the approach may anticipate glucose changes ahead of standard continuous glucose monitors (CGMs), a claim rooted in the physiology of glucose transit from bloodstream to tissue.

On a recent episode of Diabetes Dialogue, hosts Diana Isaacs, PharmD, and Natalie Bellini, DNP, are joined by Casey Halpern, MD, professor of neurosurgery and chief of stereotactic and functional neurosurgery at the University of Pennsylvania, about SynchNeuro, the company he cofounded to develop the device.

Halpern's group originally identified the underlying signal while using deep brain stimulation electrodes to study obesity and type 2 diabetes, finding that certain brain rhythms tracked blood glucose in real time regardless of electrode location. Because the signal is diffuse rather than localized, the team determined it could be captured non-invasively via scalp EEG, most reliably over the temporal lobe, an area accessible with a small sensor worn behind the ear. Device positioning has proven forgiving, with built-in signal-quality feedback guiding users to adjust placement as needed.

Accuracy data remain preliminary, drawn from EEG recordings paired with CGM readings in several dozen users over up to two weeks of continuous wear. Comparing readings synchronized to standard CGM timing, the sensor's mean absolute relative difference runs around 10% to 11%, approaching the accuracy range typically reported for interstitial fluid-based devices. Notably, Halpern reported that accuracy improves when comparing the sensor's output from 20 minutes earlier against subsequent CGM readings, suggesting the EEG signal reflects blood glucose upstream of the diffusion delay inherent to interstitial fluid sensing.

The company is enrolling additional users toward an algorithm-training cohort in the hundreds, which Halpern said internal power calculations suggest should be sufficient before pursuing regulatory clearance. Plans include an initial focus on type 2 diabetes and prediabetes, with a potential wellness-category launch preceding a full US Food and Drug Administration clinical-claims pathway. Beyond glucose, the platform is being explored for ketone monitoring, sleep apnea, and stroke risk prediction, applications that would each require separate regulatory work.

Collectively, the findings position EEG-based glucose sensing as a potential complement or alternative to interstitial fluid CGMs, particularly for patients seeking a non-invasive, intermittently wearable option, though larger validation cohorts and regulatory review will determine whether the anticipatory accuracy signal holds up at scale.

Editors’ Note: Isaacs reports disclosures with Dexcom, Abbott, Lilly, Novo Nordisk, Medtronic, Insulet, and others. Bellini reports disclosures with Abbott Diabetes Care, MannKind, Povention Bio, and others. Halpern reports being a consultant for Insightec and employment with SynchNeuro.

References

1. Kinsey N. SynchNeuro Has a Working Prototype, a New Study Launching, and a Forecast for Your Blood Sugar | Health Moonshot Update. Startup Health. May 27, 2026. Accessed August 26, 2026. https://media.startuphealth.com/p/synchneuro-has-a-working-prototype-a-new-study-launching-and-a-forecast-for-your-blood-sugar-health-