UbiStroke and HoloStroke: Multimodal and Immersive Assessment of Stroke

Overview

Stroke is a major leading cause of death and disability in the United States and stroke-related care poses a significant burden on the population. Recombinant Tissue Plasminogen Activator is the only therapy for stroke but it is utilized in less than 5% of acute strokes due to the narrow therapeutic time window for intervention, within 3 hours of the onset of symptoms. Stroke assessment is currently hindered by human analysis. Accurately diagnosing a stroke as soon as possible after the onset of symptoms requires a stroke specialist and adjunctive imaging techniques. Additionally, stroke outcome prediction is currently crude, and stroke deficit scales are generally unable to predict if a patient will do well or very poorly.

Two lines of work address that gap from opposite ends: making the deficits themselves measurable, and moving the expertise to wherever the patient is.

UbiStroke

UbiStroke develops techniques to reliably identify and quantify the degree of different stroke-related deficits, through an accurate per-patient stroke signature that will help to more precisely assess patients' deficits. The system exploits multiple sensors and generates a stroke assessment based on body posture, facial droop, pupil tracking, speech patterns, and other symptoms of stroke. We are evaluating UbiStroke’s introduction into the clinical setting to assist the clinician-performed NIHSS exam.

HoloStroke

HoloStroke grew out of UbiStroke and asks how mixed reality can extend stroke care beyond the bedside. Using the HoloLens, holograms of a patient, or of their imaging, are placed in three-dimensional space and can be transmitted to a distant site, letting a remote stroke neurologist assess and interact as if present in the same room.

The work has since branched into several directions. Holo-Stroke established that patients found remote holographic visits with a stroke provider satisfying and immersive. Holo-Stroke-CTA asked whether providers reading computed tomographic angiography for large vessel occlusion would do better in holographic space than with a conventional DICOM viewer, and found higher reliability and markedly higher satisfaction. Holo-Stroke-ii extended the approach to inpatient immersion, and Holo-PACS broadened it from stroke to radiographic scans more generally.

Funding and External Collaborations

UbiStroke and HoloStroke are a collaboration with the UCSD Stroke Center, and the UCSD Bioengineering Department, the Institute of Neural Computation, and HomniHealth. Research leading to UbiStroke has been funded by the National Science Foundation, the NSF I-Corps, the National Library of Medicine at NIH, IBM Research, and UCSD’s Office of Research Affairs.

Publications