Subhajit Chakrabarty
Link to WebpageLouisiana State University Shreveport
Project Title
Ischemic Stroke in COVID-19 Patients - Predicting Outcomes with Machine Learning.
Mentor
Karen Stokes, Louisiana State University Health Shreveport, Professor of Molecular and Cellular Physiology
Collaborator
Shashank Shekhar, University of Mississippi Medical Center
Funding Period
Startup Project (May 1, 2022 - April 30, 2024)
Abstract
Stroke is a serious complication of COVID-19. Information from brain imaging of infected patients is fragmented and the location(s) or lesion type(s) have not been fully defined, nor has their relevance to outcomes measures in these patients been properly investigated. COVID-19 is associated with a multitude of possible neurological symptoms, and it has become clear that a majority of stroke events in infected patients result from origins other than what is typically seen in non-infected individuals, i.e. ongoing large artery or small vessel disease, or a cardio-embolic event. Currently, there are no clear imaging parameters that can be used to predict duration of hospital stay, quality of life or long-term outcome of COVID-19 stroke patients. Thus, it is critical to enhance our understanding of what happens within the brain of COVID-19 stroke patients and how this relates to other clinical parameters such as coagulation panel, blood cell counts, and cardiovascular risk factors, so that we can identify imaging biomarkers that will be useful for diagnosis, outcome prediction and ultimately treatment design for these patients. Accurate identification of characteristic lesions of the brain is crucial in predicting outcome and designing treatments. Lack of literature on neurological imaging findings in COVID-19 patients, as well as human errors in reporting the imaging findings, are further hampering research.