Prediction of post-stroke aphasia treatment outcomes is significantly improved by inclusion of local resting-state fMRI measures

EasyChair Preprint no. 6434

Abstract

While the use of neural-based measures for predicting response to treatment in post-stroke aphasia (PSA) is of interest for basic science, its utility for clinical purposes is qualified by the relative difficulty and expense of collecting such measures. Thus, neural measures may be worth collecting only if they contribute unique information toward patient diagnosis or prognosis. Resting-state fMRI (rs-fMRI) is attractive because, compared to other neuroimaging approaches, the data are relatively easy to collect. While recent work indicates that local rs-fMRI measures distinguish between healthy and lesioned tissues and index domain-specific language deficits, an open question remains as to whether such measures contribute to predicting response to treatment beyond what can be predicted on the basis of demographic or behavioral measures.

Here, individuals with PSA subsequent to a single left-hemisphere stroke were treated for deficits in naming (n = 28), spelling (n = 22), or syntax (n = 14), and completed rs-fMRI scans prior to treatment. The rs-fMRI data were used to measure the fractional Amplitude of Low Frequency Fluctuations (fALFF) in the gray matter anatomical parcels of the Harvard-Oxford brain atlas. Response to treatment was first predicted using the best set of demographic and behavioral measures, and then with the addition of neural measures.

We found that for all three language domains, fALFF significantly improved predictions: the median absolute error (MAE) based on behavioral/demographic measures ranged from 11-17% across the language domains, and improved to just 1-3% when including neural measures. Similarly, 80% prediction intervals around the estimated gains on treated items narrowed from ± 22-32% to ± 4-6%, indicating that not only are predictions more precise when including fALFF, they also express more certainty.

Keyphrases: fALFF, fMRI, predicting recovery, Response to treatment, resting state

@Booklet{EasyChair:6434,
  author = {Robert Wiley and James Higgins and David Caplan and Swathi Kiran and Brenda Rapp},
  title = {Prediction of post-stroke aphasia treatment outcomes is significantly improved by inclusion of local resting-state fMRI measures},
  howpublished = {EasyChair Preprint no. 6434},

  year = {EasyChair, 2021}}