Related Experiment Video
Updated: Jan 12, 2026
Network Analysis of the Default Mode Network Using Functional Connectivity MRI in Temporal Lobe Epilepsy
Published on: August 5, 2014
Detecting temporal asymmetry after epilepsy surgery: a 3D MRI-based comparative outcome study of clinicians and lay
Rafael Denadai1, Marina Koutsodontis Machado Alvim1,2, Yeonah Kang3
1Brazilian Institute of Neuroscience and Neurotechnology (BRAINN), Universidade Estadual de Campinas, Campinas, São Paulo, Brazil.
Background:
Resective surgery through pterional approach is an effective treatment for drug-resistant temporal lobe epilepsy, but it may lead to temporal asymmetry in the craniofacial region. Nonetheless, recent systematic reviews showed that there is no gold standard method for the discrimination of a clinically visible abnormal temporal asymmetry from a normal fluctuating asymmetry. This study compares the ability of trained and untrained observers to detect temporal asymmetry and establishes a threshold for clinically detecting abnormal asymmetry.
Methods:
Standardized magnetic resonance imaging (MRI)-derived three-dimensional (3D) frontal views of adult patients who underwent temporal lobe epilepsy surgery were used to create a continuum spectrum of preoperative (n = 96) and 12-month postoperative (n = 96) craniofacial images. A panel of 32 untrained lay observers (family members and laypeople) and 32 trained clinicians (surgeons and clinical specialists) independently appraised randomly displayed 3D craniofacial soft-tissue images to assess the presence or absence of temporal asymmetry. Objective linear quantifications of temporal thickness differences were obtained from each preoperative and postoperative MRI scan to identify a potential threshold for subjective asymmetry detection. Temporal asymmetry was further categorized into severity levels I to IV based on incremental asymmetry values.
Results:
The temporal thickness difference was significantly (P < 0.001) greater in postoperative images (18.3% ± 9.6%) compared to preoperative images (4.6% ± 1.9%). As temporal asymmetry increased from Level I to IV, a significantly higher proportion of 3D craniofacial images were perceived as asymmetrical by all observer categories (P < 0.001). Both trained clinicians and untrained observers-including surgeons, clinical specialists, family members, and laypeople-demonstrated increased (P < 0.001) detection rates with increasing asymmetry severity. A temporal thickness difference exceeding 10% was clinically detected with over 90% accuracy across all observer categories (P < 0.001), establishing a 10% threshold for the clinical perception of temporal asymmetry.
