Ivan Soltesz, PhD – Stanford University, Jack Parent, MD – University of Michigan, Julie Ziobro, MD, PhD- University of Michigan and Tilo Gschwind, PhD- Stanford University
Identification of Behavioral Biomarkers in Children with DS: A Pilot Study
Special Project Funding: 2 years – $338,750
Title: Identification of Behavioral Biomarkers in Children with Dravet Syndrome: A Pilot Study
Grant Summary from the Investigators
Dravet syndrome (DS) is a severe childhood epilepsy in which seizures, developmental challenges, and heightened health risks evolve unpredictably. There are no validated biomarkers to forecast seizure risk, guide medication selection, or anticipate cognitive and motor outcomes. This leaves clinicians reliant on lengthy “trial-and-error” cycles, and families often wait months before learning whether a therapy is helping. Our project aims to address this gap by harnessing artificial-intelligence-driven behavioral phenotyping to reveal subtle movement patterns invisible to traditional observation. This automated readout could give doctors a fast, non-invasive way to see whether treatments are working. Recent preclinical work shows that, purely from subtle movement patterns, the approach can identify epileptic phenotypes and distinguish different types and doses of anti-seizure medications—even during periods when no seizures occur—signaling strong potential for patient care. By applying the same data-driven analysis in DS, we seek to uncover reliable behavioral “signatures” that reflect seizure burden, treatment responsiveness, and overall disease trajectory. A validated, non-invasive behavioral biomarker would shorten the path to the right treatment, give clinicians objective feedback on therapy efficacy, and provide researchers with a rapid outcome measure for future clinical studies. Ultimately, this work strives to augment clinical judgment with quantifiable behavioral metrics that guide personalized therapy, improving quality of life and long-term outlook for everyone affected by DS.
About the Investigators
Jack M. Parent, MD, is the William J. Herdman Professor of Neurology and Research Professor in the Michigan Neuroscience Institute at the University of Michigan Medical School, and Staff Physician at the VA Ann Arbor Medical Center. Dr. Parent received an M.D. from Yale University and completed neurology residency, clinical fellowship and postdoctoral research fellowship training at the University of California, San Francisco. In 2000, he joined the University of Michigan Department of Neurology where he co-directed the Epilepsy Division from 2007-2024 and has directed the Human Stem Cell and Gene Editing Core since 2015. His laboratory focuses on applying stem cell biology approaches to understand epileptogenic mechanisms in genetic epilepsies and related neurodevelopmental disorders. His research awards include the Dreifuss-Penry Epilepsy Award from the American Academy of Neurology, the Grass Foundation Award in Neuroscience from the American Neurological Association and the Basic Science Research Award from the American Epilepsy Society (AES). He founded and was formerly chair and then co-chair of the scientific advisory board of the Dravet Syndrome Foundation. Dr. Parent has also served as Secretary of the American Neurological Association and on the Board of Directors the AES, and is currently first vice president of the AES.
Ivan Soltesz, PhD, is the James R. Doty Professor of Neurosurgery and Neurosciences, Vice-Chair of Neurosurgery, and a member of the Wu Tsai Neurosciences Institute at Stanford University. After doctoral training in Budapest and postdoctoral work at Oxford, London, Laval, Stanford, and UT Southwestern, he founded a laboratory at the University of California, Irvine (UCI), that he later moved back to Stanford in 2015. Dr. Soltesz is internationally recognized for mechanistic insights into neuronal microcircuits, network oscillations, and cannabinoid signaling—and how their breakdown contributes to epilepsy. His laboratory integrates closed-loop in-vivo optogenetics, two-photon imaging, machine-learning-based 3D video analysis, large-scale computational modeling, and video-EEG to uncover circuit mechanisms and test therapeutic strategies. Recent work in animal models of epilepsy demonstrated that artificial-intelligence-guided analysis can detect early behavioral cues of epilepsy and medication response, findings that lay the groundwork for this project. He co-founded the inaugural Gordon Research Conference on the Mechanisms of Neuronal Synchronization and Epilepsy, authored a monograph on GABAergic microcircuits (Diversity in the Neuronal Machine, Oxford University Press), and served as editor of the book Computational Neuroscience in Epilepsy. Honors include the NINDS Javits Investigator Award, the Michael Prize in Epilepsy Research, the Basic Science Award from the American Epilepsy Society, and election as a Foreign Member of the Hungarian Academy of Sciences.
Tilo Gschwind, PhD, is an Instructor in the Department of Neurosurgery at Stanford University specializes in the development of computer vision and machine learning-guided tools to tackle major outstanding problems in neuroscience and brain disorders, with a special emphasis on epilepsy. During his PhD, he developed novel strategies to rapidly intervene upon epileptic seizures in pre-clinical models, earning the Swiss League Against Epilepsy award. As a postdoctoral research fellow at Stanford in Dr. Soltesz’ lab, he has used ML-based behavioral analysis to overcome technical barriers encountered in the diagnosis of, and drug-discovery pipeline for epilepsy, for which he received two fellowships from the Swiss National Science Foundation. His recent publication in Neuron (2023) outlined how hidden “behavioral fingerprints” can be used to accelerate the assessment of preclinical models of epilepsy and screen for treatments.
Julie Ziobro, MD, PhD, is an Assistant Professor of Pediatrics in the division of Pediatric Neurology at the University of Michigan Medical School. She received her MD and PhD degrees from Virginia Commonwealth University, where she studied cannabinoid receptor changes in models of acquired epilepsy. She completed Pediatric and Child Neurology residencies at Children’s National Hospital in Washington, DC, where she learned rodent modeling of genetic epilepsies in the lab of Dr. Judy Liu. She completed an Epilepsy fellowship at Children’s Hospital Colorado before joining the faculty at the University of Michigan. She has received K12, AES/PCDH19 Alliance, and NIH/NINDS K08 training awards under the mentorship of Dr. Jack Parent, for work exploring molecular changes in rodent models of PCDH19-Related Epilepsy. Dr. Ziobro is a co-leader of the Epilepsy Genetics clinic and the Genetic Therapeutics Action Committee at the University of Michigan, striving to provide improved diagnosis and precision therapies for patients with genetic epilepsies.