Clinical Data Animation Center · Stanford Medicine

Making brain data useful for neurological care

We develop and validate AI methods for EEG, sleep recordings and clinical records, to improve how neurological disease is measured, predicted and treated. We release the data and tools behind this work so other researchers can use them.

Led by M. Brandon Westover, MD, PhD, CDAC brings together clinicians, scientists and engineers at Stanford and collaborating institutions.

Explore our research Find data and software Join the team

Two overnight sleep EEG spectrograms with sleep-stage plots above them. The first, from a 58-year-old man, is labelled high risk; the second, from a 61-year-old man, is labelled low risk. — **Two nights of sleep, as a model reads them.** Each panel shows one person's sleep stages (top) and the frequency content of their EEG through the night (bottom). A model trained on sleep recordings placed the first night at the 95th percentile of predicted mortality risk and the second at the 5th. What sleep reveals about brain health

Who we are

The group: The Clinical Data Animation Center (CDAC), the research group of M. Brandon Westover.
Based at: Department of Neurology and Neurological Sciences, Stanford University School of Medicine.
Led by: M. Brandon Westover, MD, PhD, Director of the Stanford Epilepsy Center and of the Neurologic Artificial Intelligence Center.
Platform: CDAC runs the Brain Data Science Platform (BDSP), which shares clinical EEG, sleep and related datasets with researchers.
People: The team, affiliated investigators and collaborators

What we have built

Four examples, each with the evidence behind it. The full record is on the publications page.

EEG interpretation

A foundation model for reading clinical EEG

MORGOTH was developed and validated across multiple centres to interpret EEG automatically and comprehensively.

Lancet Digital Health, 2026 Code

Sleep and brain health

Measures of brain health from a night of sleep

Sleep EEG yields a brain age and a brain health score that track cognition and long-term outcomes.

Sleep Advances, 2026 Code

Critical care

EEG patterns and recovery after cardiac arrest

A multicentre study of 1,000 patients in coma after cardiac arrest, alongside the I-CARE EEG database that we share with other researchers.

Neurology Open Access, 2026 I-CARE data

Shared data

Clinical sleep recordings across the human lifespan

The Human Sleep Project is a multi-centre polysomnography dataset, released through BDSP for other groups to build on.

Sleep, 2026 Data on BDSP

Latest

1 Oct 2026PreprintLitBench: A Benchmark for Retrieval-Grounded Multi-Paper Evidence Synthesis in Epilepsy. is published in medRxiv
28 Sep 2026PaperAutomated Derivation of Cerebral Performance Category at Hospital Discharge After Cardiac Arrest Using Natural Language Processing and Machine Learning. is published in Neurocrit Care
22 Sep 2026PaperSleep electroencephalography-derived brain health and long-term cognitive decline in community-dwelling adults. is published in Sleep Adv
22 Sep 2026Data & codePediatric SleepNet: A Deep Learning Model for Reliable Pediatric Sleep Staging Across Developmental Stages — Trained Models and Code is published on bdsp.io
9 Sep 2026PaperCounterintuitive: reducing regression-to-the-mean worsens efficiency in simulated clinical trials. is published in Seizure
1 Sep 2026PaperPrognostic Significance of EEG Patterns in Post-Cardiac Arrest Coma: A Multicenter Retrospective Cohort Study of 1,000 Patients. is published in Neurol Open Access

All news