Christopher G. Favilla

University of Pennsylvania, PA·Medicine & Clinical Research

Strokes happen when blood flow to part of the brain is cut off, and almost everything about outcome depends on blood flow: whether a blocked artery is found fast enough to open it, how well the brain is perfused over months and years in patients with failing hearts, and whether survivors understand their own risk factors well enough to prevent the next stroke. These projects share a focus on measuring or acting on cerebral blood flow and on stroke care, using portable optical and ultrasound instruments at the bedside and in ambulances, and a personalized education platform for survivors.

Write an email

Fields

Medicine & Clinical Research

Neuroscience

Biomedical Engineering

Topics

stroke

large vessel occlusion

endovascular therapy

Grants

3 known grants, through Dec 2030

What they work on

Validate an optical algorithm for detecting large vessel occlusion by performing a 70-second speckle contrast optical spectroscopy blood flow scan on consecutive patients evaluated for acute stroke within 24 hours at three emergency departments, with urgent vascular imaging as the gold standard.
Compare the diagnostic accuracy of the optical blood flow classifier, built from a deep transformer neural network trained on cerebral blood flow waveforms, against examination-based prehospital stroke severity scales now used by emergency medical services.
Test the feasibility of portable optical blood flow imaging in ambulances by installing instruments with a large EMS provider and scanning all patients with acute focal deficit within 24 hours during prehospital stroke assessment.
Establish how left ventricular assist device implantation affects cerebral autoregulation and cerebrovascular reactivity in long-term follow-up, using transcranial Doppler ultrasonography, and relate these measures to cognition and brain atrophy over time.
Quantify how sensitive cerebral hemodynamic metrics are to device settings by monitoring cerebral blood flow during systematic LVAD pump speed titration in patients with advanced heart failure.
Evaluate in a multicenter randomized trial whether MyStroke, a platform that uses electronic health record fields to personalize educational videos, improves stroke knowledge, self-efficacy, satisfaction, and quality of life for survivors and caregivers.
Determine whether MyStroke improves medication adherence and reduces health system utilization, and compare plain versus behavioral-economics-based enhanced electronic nudges for sustaining engagement after discharge using platform analytics.
Assess digital engagement with the education platform among elderly patients and those with limited technology access to identify where the approach risks widening inequities.

More about the research

One project targets the first minutes of a stroke. Endovascular therapy, in which a catheter physically removes a clot, works well for strokes caused by a large vessel occlusion (LVO), a blockage in a major brain artery, but only a minority of stroke centers offer it. If paramedics could tell which patients have an LVO, they could drive them straight to a capable center. Emergency medical services currently rely on examination-based severity scales, which are inconsistently applied and not sensitive enough. The group validated a non-invasive optical method based on speckle contrast optical spectroscopy, which reads blood flow from the shifting speckle pattern of light scattered back out of the head, sampling fast enough and cleanly enough to resolve the shape of the blood flow waveform with each heartbeat. That waveform is distinctly abnormal in LVO. After compressing the measurement into a 70-second bedside scan, the team studied 135 emergency department patients and trained a deep transformer neural network on the waveforms; the resulting classifier detected LVO with high sensitivity and specificity and outperformed two common prehospital scales. The proposal now validates that algorithm prospectively on consecutive patients evaluated for stroke within 24 hours of onset at three Penn Medicine emergency departments, using the urgent vascular imaging already done during stroke alerts as the gold standard, and compares its accuracy with existing scales. In parallel, optical instruments will be installed in several ambulances with a large EMS partner and the 70-second scan folded into the prehospital stroke assessment, chiefly to establish feasibility and expose technical and workflow obstacles.

Why it matters

Each project aims at a concrete gap in stroke and heart failure care. A 70-second optical scan that paramedics can run could route patients with a large vessel blockage directly to hospitals able to remove the clot, shortening time to treatment. Bedside ultrasound measures of brain blood flow could give LVAD teams treatment targets, such as pump speed and blood pressure, for protecting patients from stroke and cognitive decline. A personalized video platform could help stroke survivors and caregivers understand their own risk factors, stay on prevention medications, and reduce repeat healthcare use.

Known grants

From NIH and NSF research grants. Labs often have other funding we can't see.

The impact of a personalized video-based stroke educational platform onpatient-centered and health system-centered outcomes

NIH · Feb 2026 – Dec 2030

Optical Cerebral Hemodynamic Monitoring for the Detection of Large Vessel Occlusion

NIH · Sep 2025 – Aug 2030

The impact of Left Ventricular Assist Device Implantation on Cerebral Hemodynamics and Cognition

NIH · Sep 2025 – Jul 2029

Closest labs

Maxim Mokin

University of South Florida, FL·Medicine & Clinical Research

Test whether continuous bedside optical monitoring predicts outcomes after aneurysmal subarachnoid hemorrhage, measuring cerebral blood flow and an index of cerebrovascular autoregulation with pathlength resolved diffuse correlation spectroscopy throughout acute hospitalization.

Destiny Hooper

Washington University in St Louis, MO·Medicine & Clinical Research·Early-career

Collect 40 paired arterial blood samples during endovascular thrombectomy, one from the femoral artery before the clot and one from beyond the occluded cerebral vessel via microcatheter, to sample blood adjacent to actively ischemic brain tissue.

Charles Brown

Stanford University, CA·Medicine & Clinical Research

Test whether cerebrospinal fluid markers of blood-brain barrier leak, inflammation, and neuron and astrocyte injury — including the CSF/blood albumin ratio, cytokines, neurofilament light and glial fibrillary acidic protein — are associated with delirium after aortic surgery in 180 patients with CSF drains.

The school

University of Pennsylvania: 90 labs on PI Pages, $2.2B a year on research, 750 research doctorates a year.

More about University of Pennsylvania

Reach out

A start for your email. Fill in the brackets.