Verbal memory and language fMRI in epilepsy surgery: A comparative study
Abstract number :
2.461
Submission category :
5. Neuro Imaging / 5B. Functional Imaging
Year :
2022
Submission ID :
2232868
Source :
www.aesnet.org
Presentation date :
12/4/2022 12:00:00 PM
Published date :
Nov 22, 2022, 05:28 AM
Authors :
Vasileios Kokkinos, PhD, PhD – Massachusetts General Hospital; ioannis Seimenis, PhD – Democritus University of Thrace
This is a Late Breaking abstract
Rationale: fMRI is a non-invasive and safe presurgical evaluation tool, among others used to lateralize verbal memory in patients with temporal lobe epilepsy (TLE) who are candidates for anterior temporal lobectomy (ATL). In this work we investigate whether language tasks are equivalent to verbal memory tasks in terms of lateralization, and assess the validity of performing a word recognition task during the functional scan rather than post-scan.
Methods: Thirty patients underwent verbal memory, visuospatial memory and language fMRI. Four memory tasks were used: (1) word encoding (WE), (2) word recognition (WR), (3) image encoding (IE), and (4) image recognition (IR); and three language tasks: (1) semantic description (SD), (2) reading comprehension (RD), and (3) listening comprehension (LIS). We used three common metrics, often used in the clinical context: the network spatial distribution, the maximum statistical value, and the laterality index (LI).
Results: In terms of network spatial distribution, the comparison between our verbal memory and language fMRI tasks resulted in overall poor similarity. In terms of maximum signal lateralization, the verbal memory tasks showed higher but not excellent similarity to the language tasks. The WE task LIs showed significant correlation only with the receptive language LIs of the LIS task (p = 0.016). In contrast, the LIs of the WR task were significantly correlated to the LIs of expressive language regions in the SD task (p = 0.024) and receptive language regions of the RD and LIS tasks (p = 0.015, and p = 0.019, respectively). There was no correlation between the LIs of the IE and IR tasks and any of the language tasks.
Conclusions: Our results demonstrated the value of performing a WR task during the scan for mapping purposes rather than outside the MR scanner. More importantly, our results support the inference between language and verbal memory lateralization – corresponding to word recognition/recall processes rather than encoding – when made on the basis of LI quantification.
Funding: This work was not supported by funding sources.
Neuro Imaging