Abstract #3611
Predicting recovery from stroke using baseline imaging biomarkers of structural connectome disruption
Amy Kuceyeski 1 , Babak B. Navi 2 , Hooman Kamel 2 , Norman Relkin 2 , Ashish Raj 3 , Joan Toglia 4 , Costantino Iadecola 2 , and Michael O'Dell 4
1
Radiology and the Brain and Mind Research
Institute, Weill Cornell Medical College, New York, NY,
United States,
2
Neurology
and the Brain and Mind Research Institute, Weill Cornell
Medical College, NY, United States,
3
Radiology
and the Brain and Mind Research Institute, Weill Cornell
Medical College, NY, United States,
4
Rehabilitation
Medicine, Weill Cornell Medical College, NY, United
States
This work aims to predict three aspects of post-stroke
recovery, including daily activity, cognition and basic
mobility. We compare two models, one based on patient
demographics and lesion volume and the other based on
patient demographics and structural connectome
disruption information gleaned from the Network
Modification (NeMo) Tool. Models based on the NeMo tool
had higher accuracy and lower Akaike Information
Criterion, and also provided insight into the regions
important for each of the three measured functional
domains. After thorough validation, this method could be
a valuable quantitative tool for clinicians in
developing prognoses and rehabilitation plans for
post-stroke recovery.
This abstract and the presentation materials are available to members only;
a login is required.
Join Here
