Advancing epilepsy diagnosis: A meta-analysis of artificial intelligence approaches for interictal epileptiform discharge detection.

INTRODUCTION: Interictal epileptiform discharges (IEDs) in electroencephalograms (EEGs) are an important biomarker for epilepsy. Currently, the gold standard for IED detection is the visual analysis performed by experts. However, this process is expert-biased, and time-consuming. Developing fast, accurate, and robust detection methods for IEDs based on EEG may facilitate epilepsy diagnosis. We aim to assess the performance of deep learning (DL) and classic machine learning (ML) algorithms in classifying EEG segments into IED and non-IED categories, as well as distinguishing whether the entire EEG contains IED or not. METHODS: We systematically searched PubMed, Embase, and Web of Science following PRISMA guidelines. We excluded studies that only performed the detection of IEDs instead of binary segment classification. Risk of Bias was evaluated with Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). Meta-analysis with the overall area under the Summary Receiver Operating Characteristic (SROC), sensitivity, and specificity as effect measures, was performed with R software. RESULTS: A total of 23 studies, comprising 3,629 patients, were eligible for synthesis. Eighteen models performed discharge-level classification, and 6 whole-EEG classification. For the IED-level classification, 3 models were validated in an external dataset with more than 50 patients and achieved a sensitivity of 84.9 % (95 % CI: 82.3-87.2) and a specificity of 68.7 % (95 % CI: 7.9-98.2). Five studies reported model performance using both internal validation (cross-validation) and external datasets. The meta-analysis revealed higher performance for internal validation, with 90.4 % sensitivity and 99.6 % specificity, compared to external validation, which showed 78.1 % sensitivity and 80.1 % specificity. CONCLUSION: Meta-analysis showed higher performance for models validated with resampling methods compared to those using external datasets. Only a minority of models use more robust validation techniques, which often leads to overfitting.

Transsylvian transopercular peri-central core hemispherotomy for treating epilepsy: anatomy, surgical technique, and clinical outcome.

OBJECTIVE: The objective of this study was to display the anatomical landmarks, surgical technique, and clinical outcome of transsylvian transopercular peri-central core hemispherotomy (TTPH) for treating refractory epilepsy. METHODS: From 2011 to 2023, 26 patients (12 with Rasmussen syndrome, 8 with hemimegalencephaly/cortical malformations, and 6 with hypoxic-ischemic encephalopathy; mean [range] age 11.3 years [16 months to 35 years]; 13 females; and 13 with right-side pathology) underwent TTPH. The mean (range) follow-up was 88 (14-156) months. The intradural surgical time, use and amount of blood transfusion, postoperative fever, hospital stay, weight at surgery, and seizure onset to surgery interval are reported. RESULTS: TTPH consists of 1) sylvian fissure opening, 2) coagulation of the M2 and M3 branches, 3) frontoparietal opercula removal, 4) suprainsular resection, 5) insula removal, 6) selective amygdalohippocampectomy, 7) disconnection of the posterior temporal and occipital lobes using the tentorium and falx as landmarks, 8) intraventricular callosotomy, and 9) disconnection of the basal frontal lobe. In cortical malformation, the gray-white matter interface serves as a landmark. The average intradural operating time was 7 hours 18 minutes (3 hours 33 minutes to 13 hours 45 minutes); all patients were Engel class I; and 2 patients presented with procedure-related complications (meningitis and transient abducens nerve palsy). No patient required shunt surgery or reoperation. CONCLUSIONS: TTPH offers anatomical landmarks as intraoperative guides and has achieved good seizure control and low complication rates.

Ictal SPECT in Psychogenic Nonepileptic and Epileptic Seizures.

BACKGROUND: Psychogenic nonepileptic seizures (PNES) are a common and debilitating problem in patients with epilepsy. They can be virtually indistinguishable from epileptic seizures, demanding video-electroencaphalogram monitoring, which is costly and not widely available, for differential diagnosis. Specific functional brain correlates of PNES have not been demonstrated so far. We hypothesized that PNES and epileptic seizures have distinct brain activation patterns, assessed by functional neuroimaging during ictal events of both conditions. OBJECTIVE: Compare ictal brain activation patterns of PNES and epileptic seizures using single-photon emission computerized tomography. METHODS: We prospectively assessed brain functional activation using single-photon emission computerized tomography 99mTc-ethyl cysteinate dimer in 26 patients with PNES, confirmed by trained psychiatrists in epileptology, who had their seizures induced by provocative tests compared with 22 age- and sex-matched subjects with temporal lobe epilepsy who underwent prolonged intensive video-electroencaphalogram monitoring. RESULTS: In PNES patients compared with temporal lobe epilepsy group, we found a consistent increase in regional cerebral blood flow in the right precuneus (Brodmann area 7; P = 0.003) and right posterior cingulate cortex (Brodmann area 31; P = 0.001), as well as a decrease in regional cerebral blood flow in the right amygdala (P = 0.027). CONCLUSIONS: Activation of default mode network brain areas and temporoparietal junction may be a distinct feature of ictal PNES and could be explained by a disruption between movement prediction input and sensory outcome. Such information mismatch might be the neurobiological underpinning of dissociative episodes.