Jennifer A Kim
Yale University, CT·Medicine & Clinical Research
After a traumatic brain injury (TBI), nearly one in five survivors later develops epilepsy, meaning recurring seizures. This post-traumatic epilepsy (PTE) slows recovery and adds emotional and financial burdens.
- Fields
Medicine & Clinical Research
Neuroscience
Computer Science
- Topics
post-traumatic epilepsy
traumatic brain injury
seizures
- Grants
- 1 known grant, through Feb 2031
What they work on
- Develop an algorithm that identifies post-traumatic epilepsy patients from electronic health records, and predict future PTE risk from the first seven days after TBI using structured data and natural language processing of clinical notes.
- Measure contusion (brain bruise) volume and location automatically from acute CT scans with a convolutional neural network, and test their value for predicting PTE.
- Use CNN-based segmentation of clinical MRI scans to measure cortical volume and thickness, and test their association with PTE in about 3,000 TBI survivors.
- Combine EHR, CT and MRI features into one prediction model, and test whether it outperforms models built from a single data type.
More about the research
Doctors have no reliable way to predict who will develop it. Existing risk models do not scale well, rely on data many patients lack, or require staff to pull details from medical records by hand. As a result, prevention studies are hard to run, and patients in less-resourced settings are left out.
Why it matters
Post-traumatic epilepsy affects many traumatic brain injury survivors and harms their recovery. Automated prediction from widely available records and scans could help clinicians spot high-risk patients early. It could also make prevention trials easier to run and bring risk assessment to hospitals that lack research staff.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Matthew Churpek
University of Wisconsin-Madison, WI·Medicine & Clinical Research
Improve early identification and risk stratification of infected hospital patients by applying natural language processing and deep learning to structured and free-text electronic health record data from multiple centers.
Gabriel Wardi
University of California-San Diego, CA·Medicine & Clinical Research·Early-career
Identify clinical phenotypes, or distinct patient subgroups, of sepsis that develops after hospital admission, using deep-learning analysis of electronic health record data.
Yili Zhang
Georgetown University, DC·Computer Science
Develop a large language model–based natural language processing pipeline that extracts immune-related adverse event evidence, type and grade from de-identified clinical notes, creating a reusable annotated corpus.
The school
Yale University: 73 labs on PI Pages, $1.5B a year on research, 438 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.