Lateralization of Temporal Lobe Epilepsy Based on Resting-State Functional Magnetic Resonance Imaging and Machine Learning.
Ontology highlight
ABSTRACT:
SUBMITTER: Yang Z
PROVIDER: S-EPMC4553409 | biostudies-literature | 2015
REPOSITORIES: biostudies-literature
Publications
Lateralization of Temporal Lobe Epilepsy Based on Resting-State Functional Magnetic Resonance Imaging and Machine Learning.
Yang Zhengyi Z Choupan Jeiran J Reutens David D Hocking Julia J
Frontiers in neurology 20150831
Lateralization of temporal lobe epilepsy (TLE) is critical for successful outcome of surgery to relieve seizures. TLE affects brain regions beyond the temporal lobes and has been associated with aberrant brain networks, based on evidence from functional magnetic resonance imaging. We present here a machine learning-based method for determining the laterality of TLE, using features extracted from resting-state functional connectivity of the brain. A comprehensive feature space was constructed to ...[more]
Similar Datasets
Project description:ObjectiveLocalization of focal epilepsy is critical for surgical treatment of refractory seizures. There remains a great need for noninvasive techniques to localize seizures for surgical decision-making. We investigate the use of deep learning using resting state functional magnetic resonance imaging (RS-fMRI) to identify the hemisphere of seizure onset in temporal lobe epilepsy (TLE) patients.MethodsA total of 2132 healthy controls and 32 preoperative TLE patients were studied. All participants underwent structural MRI and RS-fMRI. Healthy control data were used to generate training samples for a three-dimensional convolutional neural network (3DCNN). RS-fMRI was synthetically altered in randomly lateralized regions in the healthy control participants. The model was then trained to classify the hemisphere containing synthetic noise. Finally, the model was tested on TLE patients to assess its performance for detecting biological seizure onset zones, and gradient-weighted class activation mapping (Grad-CAM) identified the strongest predictive regions.ResultsThe 3DCNN classified healthy control hemispheres known to contain synthetic noise with 96% accuracy, and TLE hemispheres clinically identified to be seizure onset zones with 90.6% accuracy. Grad-CAM identified a range of temporal, frontal, parietal, and subcortical regions that were strong anatomical predictors of the seizure onset zone, and the resting state networks that colocalized with Grad-CAM results included default mode, medial temporal, and dorsal attention networks. Lastly, in an analysis of a subset of patients with postsurgical outcomes, the 3DCNN leveraged a more focal set of regions to achieve classification in patients with Engel Class >I compared to Engel Class I.SignificanceNoninvasive techniques capable of localizing the seizure onset zone could improve presurgical planning in patients with intractable epilepsy. We have demonstrated the ability of deep learning to identify the correct hemisphere of the seizure onset zone in TLE patients using RS-fMRI with high accuracy. This approach represents a novel technique of seizure lateralization that could improve preoperative surgical planning.
| S-EPMC9177812 | biostudies-literature
Project description:Background/aimThe main purpose of this study is to evaluate the resting state hippocampal connectivity with language areas and to correlate them with laterality index calculations on single subject basis, hence to present hippocampal lateralization for language with rs-fMRI.Materials and methodsTask based and rs-fMRI data were gathered from a total of 45 subjects in 3T scanner. BrainVoyager QX, SPM, and CONN softwares were used for data analysis. LI score of each subject was calculated and converted into normalized LI score (nLI). Intrahemispheric rs-connectivity analysis was performed between hippocampus and Broca’s regions on both sides. Correlation between these variables was measured with SPSS software.ResultsRight-TLE patients were found to have highest whereas left-TLE group were found to have lowest mean LI scores. Regarding hippocampal-lingual networks; left intrahemispheric connectivity values showed strong positive correlation with nLI values in left, right-TLE patients and healthy controls (P = 0.035, 0.014, 0.047). There were no significant correlation between right intrahemispheric connectivity values and nLI scores in all groups.ConclusionsThis study seems to depict the existence of resting state hippocampal-lingual functional network which correlates well with lateralization of language function in the left hemisphere in both temporal lobe epilepsy patients and healthy controls.
| S-EPMC7491274 | biostudies-literature
Project description:Background and purposeAlthough most studies on epilepsy have focused on the epileptogenic zone, epilepsy is a system-level disease characterized by aberrant neuronal synchronization among groups of neurons. Increasingly, studies have indicated that mesial temporal lobe epilepsy may be a network-level disease; however, few investigations have examined resting-state functional connectivity of the entire brain, particularly in patients with mesial temporal lobe epilepsy and hippocampal sclerosis. This study primarily investigated whole-brain resting-state functional connectivity abnormality in patients with mesial temporal lobe epilepsy and right hippocampal sclerosis during the interictal period.Materials and methodsWe investigated resting-state functional connectivity of 21 patients with mesial temporal lobe epilepsy with right hippocampal sclerosis and 21 neurologically healthy controls. A multivariate pattern analysis was used to identify the functional connections that most clearly differentiated patients with mesial temporal lobe epilepsy with right hippocampal sclerosis from controls.ResultsDiscriminative analysis of functional connections indicated that the patients with mesial temporal lobe epilepsy with right hippocampal sclerosis exhibited decreased resting-state functional connectivity within the right hemisphere and increased resting-state functional connectivity within the left hemisphere. Resting-state network analysis suggested that the internetwork connections typically obey the hemispheric lateralization trend and most of the functional connections that disturb the lateralization trend are the intranetwork ones.ConclusionsThe current findings suggest that weakening of the resting-state functional connectivity associated with the right hemisphere appears to strengthen resting-state functional connectivity on the contralateral side, which may be related to the seizure-induced damage and underlying compensatory mechanisms. Resting-state network-based analysis indicated that the compensatory mechanism among different resting-state networks may disturb the hemispheric lateralization.
| S-EPMC7964696 | biostudies-literature
Project description:In temporal lobe epilepsy (TLE), determining the hemispheric specialization for language before surgery is critical to preserving a patient's cognitive abilities post-surgery. To date, the major techniques utilized are limited by the capacity of patients to efficiently realize the task. We determined whether resting-state functional connectivity (rsFC) is a reliable predictor of language hemispheric dominance in right and left TLE patients, relative to controls. We chose three subregions of the inferior frontal cortex (pars orbitalis, pars triangularis, and pars opercularis) as the seed regions. All participants performed both a verb generation task and a resting-state fMRI procedure. Based on the language task, we computed a laterality index (LI) for the resulting network. This revealed that 96% of the participants were left-hemisphere dominant, although there remained a large degree of variability in the strength of left lateralization. We tested whether LI correlated with rsFC values emerging from each seed. We revealed a set of regions that was specific to each group. Unique correlations involving the epileptic mesial temporal lobe were revealed for the right and left TLE patients, but not for the controls. Importantly, for both TLE groups, the rsFC emerging from a contralateral seed was the most predictive of LI. Overall, our data depict the broad patterns of rsFC that support strong versus weak left hemisphere language laterality. This project provides the first evidence that rsFC data may potentially be used on its own to verify the strength of hemispheric dominance for language in impaired or pathologic populations.
| S-EPMC4295620 | biostudies-literature
Project description:We studied with functional magnetic resonance imaging (fMRI) differences in resting-state networks between patients with mesial temporal lobe epilepsy (MTLE) and healthy subjects. To avoid any a priori hypothesis, we use a data-driven analysis assessing differences between groups independently of structures involved. Shared and specific independent component analysis (SSICA) is an exploratory method based on independent component analysis, which performs between-group network comparison. It extracts and classifies components (networks) in those common between groups and those specific to one group. Resting fMRI data were collected from 10 healthy subjects and 10 MTLE patients. SSICA was applied multiple times with altered initializations and different numbers of specific components. This resulted in many components specific to patients and to controls. Spatial clustering identified the reliable resting-state networks among all specific components in each group. For each reliable specific network, power spectrum analysis was performed on reconstructed time-series to estimate connectivity in each group and differences between groups. Two reliable networks, corresponding to statistically significant clusters robustly detected with clustering were labeled as specific to MTLE and one as specific to the control group. The most reliable MTLE network included hippocampus and amygdala bilaterally. The other MTLE network included the postcentral gyri and temporal poles. The control-specific network included bilateral precuneus, anterior cingulate, thalamus, and parahippocampal gyrus. Results indicated that the two MTLE networks show increased connectivity in patients, whereas the control-specific network shows decreased connectivity in patients. Our findings complement results from seed-based connectivity analysis (1). The pattern of changes in connectivity between mesial temporal lobe structures and other areas may help us understand the cognitive impairments often reported in patients with MTLE.
| S-EPMC4095676 | biostudies-literature
Project description:PurposeThe purpose of this study was to investigate the local spatiotemporal consistency of spontaneous brain activity in patients with frontal lobe epilepsy (FLE).MethodEyes closed resting-state functional magnetic resonance imaging (fMRI) data were collected from 19 FLE patients and 19 age- and gender-matched healthy controls. A novel measure, named FOur-dimensional (spatiotemporal) Consistency of local neural Activities (FOCA) was used to assess the spatiotemporal consistency of local spontaneous activity (emphasizing both local temporal homogeneity and regional stability of brain activity states). Then, two-sample t test was performed to detect the FOCA differences between two groups. Partial correlations between the FOCA values and durations of epilepsy were further analyzed.Key findingsCompared with controls, FLE patients demonstrated increased FOCA in distant brain regions including the frontal and parietal cortices, as well as the basal ganglia. The decreased FOCA was located in the temporal cortex, posterior default model regions, and cerebellum. In addition, the FOCA measure was linked to the duration of epilepsy in basal ganglia.SignificanceOur study suggested that alterations of local spontaneous activity in frontoparietal cortex and basal ganglia was associated with the pathophysiology of FLE; and the abnormality in frontal and default model regions might account for the potential cognitive impairment in FLE. We also presumed that the FOCA measure had potential to provide important insights into understanding epilepsy such as FLE.
| S-EPMC5102650 | biostudies-literature
Project description:ObjectivePatients with suspected mesial temporal lobe (MTL) epilepsy typically undergo inpatient video-electroencephalography (EEG) monitoring with scalp and/or intracranial electrodes for 1 to 2 weeks to localize and lateralize the seizure focus or foci. Chronic ambulatory electrocorticography (ECoG) in patients with MTL epilepsy may provide additional information about seizure lateralization. This analysis describes data obtained from chronic ambulatory ECoG in patients with suspected bilateral MTL epilepsy in order to assess the time required to determine the seizure lateralization and whether this information could influence treatment decisions.MethodsAmbulatory ECoG was reviewed in patients with suspected bilateral MTL epilepsy who were among a larger cohort with intractable epilepsy participating in a randomized controlled trial of responsive neurostimulation. Subjects were implanted with bilateral MTL leads and a cranially implanted neurostimulator programmed to detect abnormal interictal and ictal ECoG activity. ECoG data stored by the neurostimulator were reviewed to determine the lateralization of electrographic seizures and the interval of time until independent bilateral MTL electrographic seizures were recorded.ResultsEighty-two subjects were implanted with bilateral MTL leads and followed for 4.7 years on average (median 4.9 years). Independent bilateral MTL electrographic seizures were recorded in 84%. The average time to record bilateral electrographic seizures in the ambulatory setting was 41.6 days (median 13 days, range 0-376 days). Sixteen percent had only unilateral electrographic seizures after an average of 4.6 years of recording.SignificanceAbout one third of the subjects implanted with bilateral MTL electrodes required >1 month of chronic ambulatory ECoG before the first contralateral MTL electrographic seizure was recorded. Some patients with suspected bilateral MTL seizures had only unilateral electrographic seizures. Chronic ambulatory ECoG in patients with suspected bilateral MTL seizures provides data in a naturalistic setting, may complement data from inpatient video-EEG monitoring, and can contribute to treatment decisions.
| S-EPMC4676303 | biostudies-literature
Project description:PurposeTo compare the performance of computer-automated diagnosis using functional magnetic resonance imaging (fMRI) interictal graph theory (CADFIG) to that achieved in standard clinical practice with MRI, for lateralizing the affected hemisphere in temporal lobe epilepsy (TLE).Materials and methodsInterictal resting state fMRI and high-resolution MRI were performed on 14 left and 10 right TLE patients. Functional topology measures were calculated from fMRI using graph theory, and used to lateralize the epileptogenic hemisphere using quadratic discriminant analysis. Leave-one-out cross-validation prediction accuracy of CADFIG was compared to performance based on expert manual analysis (MA) of MRI, using video EEG as the "gold standard" for focus lateralization.ResultsCADFIG correctly lateralized 95.8% (23/24) of cases, compared to 66.7% (16/24) with expert MA of MRI. Combining MA with CADFIG allowed all cases (24/24) to be correctly lateralized. CADFIG correctly identified the affected hemisphere for all patients (8/8) where MRI failed to lateralize.ConclusionCADFIG based on fMRI lateralized the affected hemisphere in TLE with superior performance compared to expert MA of MRI. These results demonstrate that functional patterns in fMRI can be used with automated machine learning for diagnostic lateralization in TLE. Addition of fMRI-based tests to existing protocols for identifying the affected hemisphere in presurgical assessment can improve diagnostic accuracy and surgical outcome in TLE.
| S-EPMC4289130 | biostudies-literature
Project description:Neuroticism, a core personality trait characterized by a tendency towards experiencing negative affect, has been reported to be higher in people with temporal lobe epilepsy (TLE) compared with healthy individuals. Neuroticism is a known predictor of depression and anxiety, which also occur more frequently in people with TLE. The purpose of this study was to identify abnormalities in whole-brain resting-state functional connectivity in relation to neuroticism in people with TLE and to determine the degree of unique versus shared patterns of abnormal connectivity in relation to elevated symptoms of depression and anxiety. Ninety-three individuals with TLE (55 females) and 40 healthy controls (18 females) from the Epilepsy Connectome Project (ECP) completed measures of neuroticism, depression, and anxiety, which were all significantly higher in people with TLE compared with controls. Resting-state functional connectivity was compared between controls and groups with TLE with high and low neuroticism using analysis of variance (ANOVA) and t-test. In secondary analyses, the same analytics were performed using measures of depression and anxiety and the unique variance in resting-state connectivity associated with neuroticism independent of symptoms of depression and anxiety identified. Increased neuroticism was significantly associated with hyposynchrony between the right hippocampus and Brodmann area (BA) 9 (region of prefrontal cortex (PFC)) (p < 0.005), representing a unique relationship independent of symptoms of depression and anxiety. Hyposynchrony of connection between the right hippocampus and BA47 (anterior frontal operculum) was associated with high neuroticism and with higher depression and anxiety scores (p < 0.05), making it a shared abnormal connection for the three measures. In conclusion, increased neuroticism exhibits both unique and shared patterns of abnormal functional connectivity with depression and anxiety symptoms between regions of the mesial temporal and frontal lobe.
| S-EPMC7483612 | biostudies-literature
Project description:PURPOSE:Temporal lobe epilepsy (TLE) affects resting state brain networks in adults. This study aims to correlate resting state functional MRI (rsMRI) signal latency in pediatric TLE patients with their laterality. METHODS:From 2006 to 2016, 26 surgical TLE patients (12 left, 14 right) with a mean age of 10.7 years (range 0.9-18) were prospectively studied. Preoperative rsMRI was obtained in patients with concordant lateralizing structural MRI, EEG, and PET studies. Standard preprocessing techniques and seed-based rsMRI analyses were performed. Additionally, the latency in rsMRI signal between each 6 mm voxel sampled was examined, compared to the global mean signal, and projected onto standard atlas space for individuals and the cohort. RESULTS:All but one of the 26 patients improved seizure frequency postoperatively with a mean follow-up of 2.9 years (range 0-7.7), with 21 patients seizure-free. When grouped for epileptogenic laterality, the latency map qualitatively demonstrated that the right TLE patients had a relatively early signal pattern, whereas the left TLE patients had a relatively late signal pattern compared to the global mean signal in the right temporal lobe. Quantitatively, the two groups had significantly different signal latency clusters in the bilateral temporal lobes (p?<?0.001). CONCLUSION:There are functional MR signal latency changes in medical refractory pediatric TLE patients. Qualitatively, signal latency in the right temporal lobe precedes the mean signal in right TLE patients and is delayed in left TLE patients. With larger confirmatory studies, preoperative rsMRI latency analysis may offer an inexpensive, noninvasive adjunct modality to lateralize pediatric TLE.
| S-EPMC5897166 | biostudies-literature