Related Experiment Video
Updated: Aug 20, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019
Topographic Mapping of Isolated Thalamic Infarcts Using Vascular and Novel Probabilistic Functional Thalamic
Maximilian Rauch1, Jan-Rüdiger Schüre2, Franziska Lieschke3
1Institute for Neuroradiology, University Hospital, Johann Wolfgang Goethe-University, Theodor Stern Kai 7, 60590, Frankfurt am Main, Germany. maximilian.rauch@kgu.de.
Purpose:
We aimed to re-evaluate the relationship between thalamic infarct (TI) localization and clinical symptoms using a vascular (VTM) and a novel functional territorial thalamic map (FTM).
Methods:
Magnetic resonance imaging (MRI) and clinical data of 65 patients with isolated TI were evaluated (female n = 23, male n = 42, right n = 23, left n = 42). A VTM depicted the known seven thalamic vascular territories (VT: inferolateral, anterolateral, inferomedial, posterior, central, anteromedian, posterolateral). An FTM was generated from a probabilistic thalamic nuclei atlas to determine six functionally defined territories (FT: anterior: memory/emotions; ventral: motor/somatosensory/language; medial: behavior/emotions/nociception, oculomotor; intralaminar: arousal/pain; lateral: visuospatial/somatosensory/conceptual and analytic thinking; posterior: audiovisual/somatosensory). Four neuroradiologists independently assigned diffusion-weighted imaging (DWI) lesions to the territories mapped by the VTM and FTM. Findings were correlated with clinical features.
Results:
The most frequent symptom was a hemisensory syndrome (58%), which was not specific for any territory. A co-occurrence of hemisensory syndrome and hemiparesis had positive predictive values (PPV) of 76% and 82% for the involvement of the inferolateral VT and ventral FT, respectively. Thalamic aphasia had a PPV of 63% each for involvement of the anterolateral VT and ventral FT. Neglect was associated with involvement of the inferolateral VT/ventral FT. Interrater reliability for the assignment of DWI lesions to the VTM was fair (κ = 0.36), but good (κ = 0.73) for the FTM.
Conclusion:
The FTM revealed a greater reproducibility for the topographical assignment of TI than the VTM. Sensorimotor hemiparesis and neglect are predictive for a TI in the inferolateral VT/ventral FT. The hemisensory syndrome alone does not allow any topographical assignment.

