Data curation and machine learning for interpretation of computed tomography (CT) images in brain injury
Brief Description of Project:
Traumatic brain injuries (TBI) affects more than 2 million patients every year in the United States, which may lead to severe physical, cognitive, and psychological disorders. Computed tomography (CT) on the human brain is one of the most widely used tools for diagnosing TBI associated with intracranial hemorrhage. In recent years, large-scale informatics analyses have shown its advantages in understanding TBI. Thus, it is appealing to perform large-scale imaging analyses on TBI cohorts as well, especially as brain trauma is a heterogenous condition with large inter-subject variations.
In this project, we will revisit four successful clinical studies (BRAINICU and MINDUSA) for which CT imaging was not a primary study consideration. We will retrieve imaging data for these participants and perform data wrangling to harmonize the data into a form suitable for machine learning. We will then applied state of the art deep learning networks to detect associations between baseline imaging data and primary study outcomes.
Strong programming skills (Python, Matlab), an interest in brain injury, and an interest in medical imaging
Nature of Supervision:
We will have weekly project-specific meetings with faculty advisors. A graduate student mentor will provide regular guidance / support. There will be whole group weekly update meetings for the broader lab group.
A Brief Research Plan (period is for 10 weeks):
The primary focus of this project is on data wrangling (curation, validation, quality assurance, and integration of imaging data from multiple existing clinical research projects) [estimated duration 6 weeks]. The secondary objective is to apply deep learning models to capture possible associations between baseline imaging and outcome variables in the clinical studies [3 weeks]. A substantial level of effort will be applied toward characterization, documentation, and presentation of the methods developed and results discovered through technical writing and oral presentations [1 week].
Number of Open Slots: 1
Name: Bennett Landman
Title: Associate Professor
Department: Electrical Engineering
Campus Address: FGH 372
Phone: (615) 322-2338