Development and prospective clinical validation of a convolutional neural network for automated detection and segmentation of focal cortical dysplasias.

PURPOSE: Focal cortical dysplasias (FCDs) are a leading cause of drug-resistant epilepsy. Early detection and resection of FCDs have favorable prognostic implications for postoperative seizure freedom. Despite advancements in imaging methods, FCD detection remains challenging. House et al. (2021) introduced a convolutional neural network (CNN) for automated FCD detection and segmentation, achieving a sensitivity of 77.8%. However, its clinical applicability was limited due to a low specificity of 5.5%. The objective of this study was to improve the CNN's performance through data-driven training and algorithm optimization, followed by a prospective validation on daily-routine MRIs. MATERIAL AND METHODS: A dataset of 300 3 T MRIs from daily clinical practice, including 3D T1 and FLAIR sequences, was prospectively compiled. The MRIs were visually evaluated by two neuroradiologists and underwent morphometric assessment by two epileptologists. The dataset included 30 FCD cases (11 female, mean age: 28.1 ± 10.1 years) and a control group of 150 normal cases (97 female, mean age: 32.8 ± 14.9 years), along with 120 non-FCD pathological cases (64 female, mean age: 38.4 ± 18.4 years). The dataset was divided into three subsets, each analyzed by the CNN. Subsequently, the CNN underwent a two-phase-training process, incorporating subset MRIs and expert-labeled FCD maps. This training employed both classical and continual learning techniques. The CNN's performance was validated by comparing the baseline model with the trained models at two training levels. RESULTS: In prospective validation, the best model trained using continual learning achieved a sensitivity of 90.0%, specificity of 70.0%, and accuracy of 72.0%, with an average of 0.41 false positive clusters detected per MRI. For FCD segmentation, an average Dice coefficient of 0.56 was attained. The model's performance improved in each training phase while maintaining a high level of sensitivity. Continual learning outperformed classical learning in this regard. CONCLUSIONS: Our study presents a promising CNN for FCD detection and segmentation, exhibiting both high sensitivity and specificity. Furthermore, the model demonstrates continuous improvement with the inclusion of more clinical MRI data. We consider our CNN a valuable tool for automated, examiner-independent FCD detection in daily clinical practice, potentially addressing the underutilization of epilepsy surgery in drug-resistant focal epilepsy and thereby improving patient outcomes.

Predicting the Outcome of Epilepsy Surgery by Covariance Pattern Analysis of Ictal Perfusion SPECT.

Previous studies on the utility of specific perfusion patterns in ictal brain perfusion SPECT for predicting the outcome of temporal lobe epilepsy surgery used qualitative visual pattern classification, semiquantitative region-of-interest analysis, or conventional univariate voxel-based testing, which are limited by intra- and interrater variability or low sensitivity to capture functional interactions among brain regions. The present study performed covariance pattern analysis of ictal perfusion SPECT using the scaled subprofile model for unbiased identification of predictive covariance patterns. Methods: The study retrospectively included 18 responders to temporal lobe epilepsy surgery (Engel I-A at 12 mo follow-up) and 18 nonresponders (≥Engel I-B). Ictal SPECT images were analyzed with the scaled subprofile model masked to group membership for unbiased identification of the 16 covariance patterns explaining the highest proportion of variance in the whole dataset. Individual expression scores of the covariance patterns were evaluated for predicting seizure freedom after temporal lobe surgery by receiver-operating-characteristic analysis. Kaplan-Meier analysis including all available follow-up data (up to 60 mo after surgery) was also performed. Results: Among the 16 covariance patterns only 1 showed a different expression between responders and nonresponders (P = 0.03). This favorable ictal perfusion pattern resembled the typical ictal perfusion pattern in temporomesial epilepsy. The expression score of the pattern provided an area of 0.744 (95% CI, 0.577-0.911, P = 0.004) under the receiver-operating-characteristic curve. Kaplan-Meier analysis revealed a statistical trend toward longer seizure freedom in patients with positive expression score (P = 0.06). The median estimated seizure-free time was 48 mo in patients with positive expression score versus 6 mo in patients with negative expression score. Conclusion: The expression of the favorable ictal perfusion pattern identified by covariance analysis of ictal brain perfusion SPECT provides independent (from demographic and clinical variables) information for the prediction of seizure freedom after temporal lobe epilepsy surgery. The expression of this pattern is easily computed for new ictal SPECT images and, therefore, might be used to support the decision for or against temporal lobe surgery in clinical patient care.

No Evidence to Favor 99mTc-HMPAO or 99mTc-ECD for Ictal Brain Perfusion SPECT for Identification of the Seizure Onset Zone.

PURPOSE: Ictal brain perfusion SPECT with the tracer 99mTc-HMPAO or 99mTc-ECD is widely used for identification of the epileptic seizure onset zone (SOZ) in presurgical evaluation if standard pointers are uncertain or inconsistent. For both tracers, there are theoretical arguments to favor it over the other for this task. The aim of this study was to compare the performance of ictal brain perfusion SPECT between 99mTc-HMPAO and 99mTc-ECD in a rather large patient sample. PATIENTS AND METHODS: The study retrospectively included 196 patients from clinical routine in whom ictal perfusion SPECT had been performed with stabilized 99mTc-HMPAO (n = 110) or 99mTc-ECD (n = 86). Lateralization and localization of the SOZ were obtained by the consensus of 2 independent readers based on visual inspection of the SPECT images. RESULTS: The 99mTc-HMPAO group and the 99mTc-ECD group were well matched with respect to age, sex, age at first seizure, duration of disease, seizure frequency, history of previous brain surgery, and findings of presurgical MRI. The proportion of lateralizing ictal SPECT did not differ significantly between 99mTc-HMPAO and 99mTc-ECD (65.5% vs 72.1%, P = 0.36). Sensitivity of ictal perfusion SPECT (independent of the tracer) for correct localization of the SOZ in 62 patients with temporal lobe epilepsy and at least worthwhile improvement (Engel scale ≤ III) 12 months after temporal epilepsy surgery was 63%. CONCLUSIONS: This study does not provide evidence to favor 99mTc-HMPAO or 99mTc-ECD for identification of the SOZ by ictal perfusion SPECT.

Automated detection and segmentation of focal cortical dysplasias (FCDs) with artificial intelligence: Presentation of a novel convolutional neural network and its prospective clinical validation.

PURPOSE: Focal cortical dysplasias (FCDs) represent one of the most frequent causes of pharmaco-resistant focal epilepsies. Despite improved clinical imaging methods over the past years, FCD detection remains challenging, as FCDs vary in location, size, and shape and commonly blend into surrounding tissues without clear definable boundaries. We developed a novel convolutional neural network for FCD detection and segmentation and validated it prospectively on daily-routine MRIs. MATERIAL AND METHODS: The neural network was trained on 201 T1 and FLAIR 3 T MRI volume sequences of 158 patients with mainly FCDs, regardless of type, and 7 focal PMG. Non-FCD/PMG MRIs, drawn from 100 normal MRIs and 50 MRIs with non-FCD/PMG pathologies, were added to the training. We applied the algorithm prospectively on 100 consecutive MRIs of patients with focal epilepsy from daily clinical practice. The results were compared with corresponding neuroradiological reports and morphometric MRI analyses evaluated by an experienced epileptologist. RESULTS: Best training results reached a sensitivity (recall) of 70.1 % and a precision of 54.3 % for detecting FCDs. Applied on the daily-routine MRIs, 7 out of 9 FCDs were detected and segmented correctly with a sensitivity of 77.8 % and a specificity of 5.5 %. The results of conventional visual analyses were 33.3 % and 94.5 %, respectively (3/9 FCDs detected); the results of morphometric analyses with overall epileptologic evaluation were both 100 % (9/9 FCDs detected) and thus served as reference. CONCLUSION: We developed a 3D convolutional neural network with autoencoder regularization for FCD detection and segmentation. Our algorithm employs the largest FCD training dataset to date with various types of FCDs and some focal PMG. It provided a higher sensitivity in detecting FCDs than conventional visual analyses. Despite its low specificity, the number of false positively predicted lesions per MRI was lower than with morphometric analysis. We consider our algorithm already useful for FCD pre-screening in everyday clinical practice.

Influence of stiripentol on perampanel serum levels.

Stiripentol is an orphan drug successfully used in combination with valproate and clobazam in the treatment of Dravet syndrome. Perampanel has also been added by experts. In this retrospective study, we investigated the influence of stiripentol on perampanel serum levels by using the doses and corresponding perampanel serum levels of 10 patients. The impact of stiripentol on the perampanel serum levels was significant as shown in a linear regression analysis, with perampanel serum levels increasing linearly with the stiripentol doses. We conclude that dose adjustments of stiripentol and perampanel when administered together, should be done carefully to avoid side effects or even severe intoxication. Hence, perampanel serum level monitoring seems advisable.

Use of the mixed reality tool "VSI Patient Education" for more comprehensible and imaginable patient educations before epilepsy surgery and stereotactic implantation of DBS or stereo-EEG electrodes.

PURPOSE: It is unknown which patient education strategy before epilepsy surgery or stereotactic electrode implantation is best for patients. This prospective and randomized clinical study investigates whether the use of the mixed reality tool "VSI Patient Education" (VSI PE) running on HoloLens® glasses is superior to the use of a rubber brain model as a 3-dimensional tool for patient education before epilepsy surgery and stereotactic electrode implantation. MATERIAL AND METHODS: 17 patients with indication for epilepsy surgery or stereotactic electrode implantation were included in the study and randomized into two groups. All patients were informed with both comparative tools VSI PE (apoQlar®) and a rubber brain model (3B Scientific®) in a chronological order depending on group assignment. Afterwards, the patient and, if present, a relative (12) each filled out a questionnaire. For statistical analysis, Wilcoxon rank-sum tests were performed. RESULTS: Patients found their patient education highly significantly more comprehensible (p = 0.001**, r = 0.84) and almost significantly more imaginable (p=0.020, r = 0.57), when their doctor used VSI PE compared to the rubber brain model. The patients felt significantly less anxious as a result of VSI PE (p = 0.008*, r = 0.64). Highly significantly more patients chose VSI PE as the preferred patient education tool (p < 0.001**, r = 0.91), and almost significantly more patients decided VSI PE to be the future standard tool (p = 0.020, r = 0.56). Significantly more relatives chose VSI PE as the preferred patient education tool (p = 0.004*, r = 0.83), and significantly more relatives decided VSI PE to be the future standard tool (p = 0.002*, r = 0.91). CONCLUSION: VSI Patient Education is a promising new mixed reality tool for informing patients before epileptic surgery or stereotactic electrode implantation in order to enhance comprehension and imagination and reduce fear and worries. It might strengthen patient commitment and have a positive influence on patients' decisions in favor of medically indicated surgical operations.

Complex sexual behaviors during sleep as a manifestation of epilepsy: a case series.

STUDY OBJECTIVES: Complex sexual behavior during sleep (CSBS) is a well described clinical entity in nonrapid eye movement (NREM) sleep parasomnias (i.e. sexsomnia). We report a retrospective case series of CSBS as clinical manifestation of epileptic seizures and compare them with the semiology of sexsomnia. METHODS: Video-electroencephalopraphy (EEG)-monitoring data of patients with epileptic and nonepileptic paroxysmal events from one tertiary epilepsy center between 2013 and 2016 were retrospectively reviewed. Clinical features and presurgical, electroclinical, and follow-up data are presented and then discussed in the context of other published cases. RESULTS: From 4,629 patients, 6 patients had CSBS. EEG, single photon emission computed tomography (SPECT), magnetic resonance imaging (MRI), and histopathology confirmed an epileptic origin in four female patients, with temporal or frontal seizures. Two male patients had sexsomnia. None of the epilepsy patients had parasomnias. Clinical criteria to differentiate epileptic from parasomnic CSBS were: events also occurred out of wakefulness; current presence of additional nonsexual manifestations of epilepsy; sexual behavior only as part of a broad spectrum of emotional and motor automatisms; stereotyped behavior pattern without modulability by bystanders; unarousability during the event; no completion of sexual intercourse. The accuracy of the clinical diagnosis was improved by the development of an algorithm comparing patients' fulfillment of the criteria of epilepsy versus parasomnia. CONCLUSIONS: In our cohort, CSBS was a rare ictal phenomenon in temporal or frontal seizures. Symptomatological similiarities with sexsomnia might be explained by the same phylogenetically primitive "central pattern generator" manifesting in ictal CSBS by activation and in sexsomnia by disinhibition. Ictal CSBS should be considered in the differential diagnosis of sexsomnia.

Impact of focal cortical dysplasia Type IIIa on seizure outcome following anterior mesial temporal lobe resection for the treatment of epilepsy.

OBJECTIVE Temporal lobe epilepsy (TLE) is the most common type of pharmacoresistant focal epilepsy, for which anterior mesial temporal lobe resection (AMTLR) is a treatment option. Focal cortical dysplasia Type IIIa (FCD IIIa), a developmental lesion resulting from defects in neuronal formation and migration into the temporal pole (FCD I) combined with hippocampal sclerosis (HS), can be a neuropathological finding. In this study, the authors investigate the impact of FCD IIIa on seizure outcome in patients with TLE who underwent AMTLR. METHODS The authors performed a retrospective analysis of all patients with TLE who underwent AMTLR at their institution between June 2011 and April 2014. Histopathological analysis was used to determine whether patients had HS together with FCD I (FCD IIIa) or HS alone. The groups were compared with regard to age, sex, years of epilepsy, and seizure outcome using the Engel classification. RESULTS A total of 51 patients with TLE underwent AMTLR at the authors' institution. FCD IIIa was diagnosed in 13 cases. The patients experienced seizures for a mean duration of 31.1 years. The mean length of follow-up after the procedure was 18 months. All patients with FCD IIIa had a favorable seizure outcome (Engel Class I or II) compared with 71% of the patients with no pathological findings in the temporal pole (p < 0.01). CONCLUSIONS Patients with histopathologically proven FCD IIIa had a significantly better seizure outcome after AMTLR than patients with HS alone. Further effort should be made during presurgical evaluation to detect FCD IIIa so that the most suitable resection technique can be chosen and postoperative seizure outcome can be predicted for patient counseling.

Octreotide: The IIH therapy beyond weight loss, carbonic anhydrase inhibitors, lumbar punctures and surgical/interventional treatments.

OBJECTIVE: Idiopathic intracranial hypertension (IIH) is characterized by elevated intracranial pressure without a space-occupying cerebral lesion, venous sinus thrombosis or hydrocephalus and with normally composited cerebrospinal fluid (CSF). Main symptoms are headache, sight disturbances and potential visual impairment. Weight loss, carbonic anhydrase inhibitors, lumbar punctures with CSF drain, CSF shunting, optic nerve sheath fenestration, and venous sinus stenting are common IIH therapies. Octreotide, a synthetic somatostatin analogue, also effectively suspends IIH symptoms. Here, treatment with octreotide on five so far inefficiently treated IIH patients is presented and discussed. METHODS: Five female patients with IIH and a history of failed therapy regimes were treated with octreotide, to be administered everyday subcutaneously for six months with identified doses high enough to suspend all clinical IIH symptoms. After tapering for two months, the further clinical course was to be monitored. RESULTS: All patients were IIH symptom-free under octreotide. After tapering, one patient remained IIH symptom-free; one patient became IIH symptom-free under intramuscular octreotide after failure of former therapy regimes; one patient became IIH symptom-free on low-dose carbonic anhydrase inhibitors; one patient had an allergic reaction and paused octreotide, after successful desensitization, tooth ache developed, forcing octreotide tapering; one patient had another shunt revision alleviating IIH symptoms. CONCLUSION: We confirmed that a) clinical IIH symptoms are suspended during octreotide exposure; b) 6-month administration can sustainably abolish IIH symptoms; c) desensitization is possible for octreotide allergy. When IIH symptoms reoccur after limited-time octreotide administration, re-application of formerly ineffective carbonic anhydrase inhibitors can suspend IIH symptoms. Intramuscular octreotide is a promising long-term therapy option.

Morphometric MRI analysis enhances visualization of cortical tubers in tuberous sclerosis.

PURPOSE: Focal cortical dysplasias (FCD) type IIb and cortical tubers in tuberous sclerosis complex (TSC) are histopathologically similar and are both epileptogenic lesions frequently causing pharmacoresistant epilepsies. Morphometric analysis of T1- and T2-weighted MRI volume data sets can enhance visualization of FCD. Here, we retrospectively investigated whether morphometric MRI analysis is of equal benefit for visualizing cortical tubers. MATERIALS AND METHODS: Morphometric analysis was applied to T1- and partly also T2-weighted 1.5T or 3T MRI volume data sets of 15 TSC patients using a fully automated MATLAB(®) script (i.e. MAP07) commonly used for FCD detection. In this study, focus was on the most sensitive of the resulting morphometric feature maps (i.e. the 'junction image') which highlights blurring of the gray-white matter junction in comparison to a normal database. The visualization of tubers in these 'junction images' was quantitatively compared with that in conventional MR sequences. RESULTS: In all patients, morphometric analysis visualized almost all tubers detected in the normal MRI, and additionally highlighted on average 23% (range 3-50%) more tubers which were not detected by visual analysis of the conventional MR sequences. When T2 volume data sets from a 3T scanner were available for postprocessing, the rate of additionally detected tubers increased to 29% on average. These formerly overlooked tubers were usually smaller than the tubers already found in the conventional MRI. CONCLUSION: Morphometric analysis of MRIs in TSC can highlight cortical tubers which are likely to be overlooked in conventional MRI sequences alone. Additionally detected tubers may be of potential importance for both presurgical evaluation and initial diagnosis of TSC.

Deep brain stimulation of anterior nucleus thalami disrupts sleep in epilepsy patients.

In view of the regulatory function of the thalamus in the sleep-wake cycle, the impact of deep brain stimulation (DBS) of the anterior nucleus thalami (ANT) on sleep was assessed in a small consecutive cohort of epilepsy patients with standardized polysomnography (PSG). In nine patients treated with ANT-DBS (voltage 5 V, frequency 145 Hz, cyclic mode), the number of arousals during stimulation and nonstimulation periods, neuropsychiatric symptoms (npS), and seizure frequency were determined. Electroclinical arousals were triggered in 14.0 to 67.0% (mean 42.4 ± SD 16.8%) of all deep brain stimuli. Six patients reported npS. Nocturnal DBS voltages were reduced in eight patients (one patient without npS refused) and PSGs were repeated. Electroclinical arousals occurred between 1.4 and 6.7 (mean 3.3 ± 1.7) times more frequently during stimulation periods compared to nonstimulation periods; the number of arousals positively correlated with the level of DBS voltage (range 1 V to 5 V) (Spearman's rank coefficient 0.53121; p < 0.05). No patient experienced seizure deterioration and four patients reported remission of npS. This case-cohort study provides evidence that ANT-DBS interrupts sleep in a voltage-dependent manner, thus putatively resulting in an increase of npS. Reduction of nocturnal DBS voltage seems to lead to improvement of npS without hampering efficacy of ANT-DBS.

Comparison of morphometric analysis based on T1- and T2-weighted MRI data for visualization of focal cortical dysplasia.

PURPOSE: Focal cortical dysplasias (FCD) are highly epileptogenic lesions frequently accounting for pharmaco-resistant focal epilepsy. Visual MRI analysis combined with morphometric analysis of T1-weighted MRI data was shown to be of higher diagnostic sensitivity in detecting and delineating FCD than conventional visual analysis alone. Here we investigate whether morphometric analysis of T2-weighted MRI volume data sets is of equal benefit or perhaps more helpful for visualizing FCD. MATERIALS AND METHODS: Morphometric analysis was applied to T1- and T2-weighted MRI volume data sets of 20 epilepsy patients with FCD using a fully automated MATLAB script with scanner- and sequence-specific normal databases for T1 and T2 images. For each modality, a new feature map (i.e., 'junction image') highlighting the FCD-typical blurring of the gray-white matter junction and quantifying this feature in comparison to the normal database in terms of z-scores was calculated. The resulting T1 and T2 'junction images' were compared for conspicuity and recognizability of the FCD both qualitatively by visual assessment and quantitatively by analysis of the mean z-scores inside and outside the lesions. RESULTS: In 80% of the cases, the FCD presented with higher contrast and/or clearer delineation in the T2 than in the T1 'junction images' and were thus easier to recognize in these images. The quantitative analysis supported this impression: in 95% of cases, the ratio of mean z-scores inside and outside the FCD was higher in T2- than in T1-based 'junction images'. For the T2 'junction images', this ratio amounted to 8.7 on average and was thus more than twice as high as the corresponding T1 result of 3.7 (p<.003). CONCLUSION: Concerning visualization of FCD by highlighting blurring of the gray-white matter junction, the results of the present study indicate that morphometric analysis of T2-weighted MRI data on average is superior to T1-based morphometry.

Presurgical functional transcranial Doppler sonography (fTCD) with intravenous echo enhancing agent SonoVue enables determination of language lateralization in epilepsy patients with poor temporal bone windows.

Presurgical determination of language lateralization is important for planning and outcome estimation of neurosurgical interventions in patients with drug-refractory epilepsy. Functional transcranial Doppler sonography (fTCD) provides an established measure for language lateralization using the temporal bone windows for continuous recording of the cerebral blood flow velocity (CBFV) in both middle cerebral arteries (MCAs). However, because of insufficient temporal bone windows, fTCD cannot be applied properly in every patient. Here, we established stable and sufficient CBFV signals in both MCAs using continuous intravenous application of echo-enhancing agent SonoVue in 7 of 10 patients with poor temporal bone windows and were thus able to determine language lateralization. We conclude that the application of SonoVue can solve one principal disadvantage of fTCD and improves the applicability of the technique as a presurgical functional language lateralization procedure.