Clinical validation of automated hippocampal segmentation in temporal lobe epilepsy

Artificial intelligence in epilepsy - applications and pathways to the clinic

Artificial intelligence (AI) is rapidly transforming health care, and its applications in epilepsy have increased exponentially over the past decade. Integration of AI into epilepsy management promises to revolutionize the diagnosis and treatment of this complex disorder. However, translation of AI into neurology clinical practice has not yet been successful, emphasizing the need to consider progress to date and assess challenges and limitations of AI. In this Review, we provide an overview of AI applications that have been developed in epilepsy using a variety of data modalities: neuroimaging, electroencephalography, electronic health records, medical devices and multimodal data integration. For each, we consider potential applications, including seizure detection and prediction, seizure lateralization, localization of the seizure-onset zone and assessment for surgical or neurostimulation interventions, and review the performance of AI tools developed to date. We also discuss methodological considerations and challenges that must be addressed to successfully integrate AI into clinical practice. Our goal is to provide an overview of the current state of the field and provide guidance for leveraging AI in future to improve management of epilepsy.

Automated protocols for delineating human hippocampal subfields from 3 Tesla and 7 Tesla magnetic resonance imaging data

Researchers who study the human hippocampus are naturally interested in how its subfields function. However, many researchers are precluded from examining subfields because their manual delineation from magnetic resonance imaging (MRI) scans (still the gold standard approach) is time consuming and requires significant expertise. To help ameliorate this issue, we present here two protocols, one for 3T MRI and the other for 7T MRI, that permit automated hippocampus segmentation into six subregions, namely dentate gyrus/cornu ammonis (CA)4, CA2/3, CA1, subiculum, pre/parasubiculum, and uncus along the entire length of the hippocampus. These protocols are particularly notable relative to existing resources in that they were trained and tested using large numbers of healthy young adults (n = 140 at 3T, n = 40 at 7T) whose hippocampi were manually segmented by experts from MRI scans. Using inter-rater reliability analyses, we showed that the quality of automated segmentations produced by these protocols was high and comparable to expert manual segmenters. We provide full open access to the automated protocols, and anticipate they will save hippocampus researchers a significant amount of time. They could also help to catalyze subfield research, which is essential for gaining a full understanding of how the hippocampus functions.

Mapping hippocampal and thalamic atrophy in epilepsy: A 7-T magnetic resonance imaging study

OBJECTIVE

Epilepsy patients are often grouped together by clinical variables. Quantitative neuroimaging metrics can provide a data-driven alternative for grouping of patients. In this work, we leverage ultra-high-field 7-T structural magnetic resonance imaging (MRI) to characterize volumetric atrophy patterns across hippocampal subfields and thalamic nuclei in drug-resistant focal epilepsy.

METHODS

Forty-two drug-resistant epilepsy patients and 13 controls with 7-T structural neuroimaging were included in this study. We measured hippocampal subfield and thalamic nuclei volumetry, and applied an unsupervised machine learning algorithm, Latent Dirichlet Allocation (LDA), to estimate atrophy patterns across the hippocampal subfields and thalamic nuclei of patients. We studied the association between predefined clinical groups and the estimated atrophy patterns. Additionally, we used hierarchical clustering on the LDA factors to group patients in a data-driven approach.

RESULTS

In patients with mesial temporal sclerosis (MTS), we found a significant decrease in volume across all ipsilateral hippocampal subfields (false discovery rate-corrected p [pFDR] < .01) as well as in some ipsilateral (pFDR < .05) and contralateral (pFDR < .01) thalamic nuclei. In left temporal lobe epilepsy (L-TLE) we saw ipsilateral hippocampal and some bilateral thalamic atrophy (pFDR < .05), whereas in right temporal lobe epilepsy (R-TLE) extensive bilateral hippocampal and thalamic atrophy was observed (pFDR < .05). Atrophy factors demonstrated that our MTS cohort had two atrophy phenotypes: one that affected the ipsilateral hippocampus and one that affected the ipsilateral hippocampus and bilateral anterior thalamus. Atrophy factors demonstrated posterior thalamic atrophy in R-TLE, whereas an anterior thalamic atrophy pattern was more common in L-TLE. Finally, hierarchical clustering of atrophy patterns recapitulated clusters with homogeneous clinical properties.

SIGNIFICANCE

Leveraging 7-T MRI, we demonstrate widespread hippocampal and thalamic atrophy in epilepsy. Through unsupervised machine learning, we demonstrate patterns of volumetric atrophy that vary depending on disease subtype. Incorporating these atrophy patterns into clinical practice could help better stratify patients to surgical treatments and specific device implantation strategies.

Man versus machine: Automatic pedicle screw planning using registration-based techniques compared with manual screw planning for thoracolumbar fusion surgeries

OBJECTIVE

This study evaluates the precision of a commercially available spine planning software in automatic spine labelling and screw-trajectory proposal.

METHODS

The software uses automatic segmentation and registration of the vertebra to generate screw proposals. 877 trajectories were compared. Four neurosurgeons assessed suggested trajectories, performed corrections, and manually planned pedicle screws. Additionally, automatic identification/labelling was evaluated.

RESULTS

Automatic labelling was correct in 89% of the cases. 92.9% of automatically planned trajectories were in accordance with G&R grade A + B. Automatic mode reduced the time spent planning screw trajectories by 7 s per screw to 20 s per vertebra. Manual mode yielded differences in screw-length between surgeons (largest distribution peak: 5 mm), automatic in contrast at 0 mm. The size of suggested pedicle screws was significantly smaller (largest peaks in difference between 0.5 and 3 mm) than the surgeon's choice.

CONCLUSION

Automatic identification of vertebrae works in most cases and suggested pedicle screw trajectories are acceptable. So far, it does not substitute for an experienced surgeon's assessment.

Development of an objective method to quantify hippocampal dentation

Hippocampal dentation (HD) refers to a series of ridges (dentes) seen on the inferior aspect of the hippocampus. The degree of HD varies dramatically across healthy individuals, and hippocampal pathology may result in loss of HD. Existing studies show associations between HD and memory performance in healthy adults as well as temporal lobe epilepsy (TLE) patients. However, until now studies relied on visual assessment of HD as no objective methods to quantify HD have been described. In this work, we describe a method to objectively quantify HD by transforming the characteristic 3D surface morphology of HD into a simplified 2D plot for which area under the curve (AUC) was calculated. This was applied to T1w scans of 59 TLE subjects, each with one epileptic hippocampus and one normal appearing hippocampus. Results showed that AUC significantly correlated with the number of dentes based on visual inspection (p < .05) and correctly sorted a set of hippocampi from least to most prominently dentated. Intra- and inter-rater reliability was nearly perfect (ICC ≥ 0.99). AUC values were significantly lower in epileptic hippocampi compared to contralateral hippocampi (p = .00019), consistent with previously published findings. In the left TLE group, the AUC values from the contralateral hippocampi showed a positive trend (p = .07) with verbal memory acquisition scores but was not statistically significant. The proposed approach is the first objective, quantitative measurement of dentation described in the literature. The AUC values numerically capture the complex surface contour information of HD and will enable future study of this interesting morphologic feature.

Loss of hippocampal dentation in hippocampal sclerosis and its relationship to memory dysfunction

OBJECTIVE

Hippocampal dentation (HD) is a "tooth-like" morphological feature observed on the inferior aspect of the human hippocampus. It has been found that HD varies dramatically in healthy adults and is positively associated with verbal and visual memory. In this work, we evaluate the loss of HD and its association to memory dysfunction in patients with temporal lobe epilepsy who have hippocampal sclerosis (HS).

METHODS

58 unilateral HS patients with neuropsychological data were identified from a retrospective database. T1w MPRAGE images (~1mm resolution) were upsampled to 0.25mm and were processed using ASHS software to obtain ultra high resolution segmentations and 3D renderings. Dentes were counted on the epileptic and contralateral sides, and associations were tested between dentation on the epileptic versus contralateral sides and measures of verbal and visuospatial memory with respect to the dominant versus non-dominant hemisphere.

RESULTS

The median number of dentes in epileptic hippocampi was significantly lower than in contralateral hippocampi (p<0.0001). Among cases with HS in the dominant hemisphere, verbal memory was significantly correlated with contralateral non-dominant hemisphere dentation (r = 0.45, p = 0.02). Similarly, among cases of HS in the non-dominant hemisphere, visual memory was significantly correlated with contralateral dominant hemisphere dentation (r = 0.50, p = 0.03). All other analyses were not significant.

SIGNIFICANCE

This is the first study characterizing dentation in TLE patients with HS and its memory correlates. There is marked loss of dentation in sclerotic hippocampi compared to the unaffected contralateral hippocampi. Material-specific measures of memory performance are paradoxically correlated with dentation contralateral to the side with HS, suggesting that contralateral functional capacity explains some of the variation in memory across TLE patients. Hippocampal dentation is an important variable to consider in understanding memory loss in TLE.

MRI-Based Machine Learning Prediction Framework to Lateralize Hippocampal Sclerosis in Patients With Temporal Lobe Epilepsy

BACKGROUND AND OBJECTIVES

MRI fails to reveal hippocampal pathology in 30% to 50% of temporal lobe epilepsy (TLE) surgical candidates. To address this clinical challenge, we developed an automated MRI-based classifier that lateralizes the side of covert hippocampal pathology in TLE.

METHODS

We trained a surface-based linear discriminant classifier that uses T1-weighted (morphology) and T2-weighted and fluid-attenuated inversion recovery (FLAIR)/T1 (intensity) features. The classifier was trained on 60 patients with TLE (mean age 35.6 years, 58% female) with histologically verified hippocampal sclerosis (HS). Images were deemed to be MRI negative in 42% of cases on the basis of neuroradiologic reading (40% based on hippocampal volumetry). The predictive model automatically labeled patients as having left or right TLE. Lateralization accuracy was compared to electroclinical data, including side of surgery. Accuracy of the classifier was further assessed in 2 independent TLE cohorts with similar demographics and electroclinical characteristics (n = 57, 58% MRI negative).

RESULTS

The overall lateralization accuracy was 93% (95% confidence interval 92%-94%), regardless of HS visibility. In MRI-negative TLE, the combination of T2 and FLAIR/T1 intensities provided the highest accuracy in both the training (84%, area under the curve [AUC] 0.95 ± 0.02) and validation (cohort 1 90%, AUC 0.99; cohort 2 76%, AUC 0.94) cohorts.

DISCUSSION

This prediction model for TLE lateralization operates on readily available conventional MRI contrasts and offers gain in accuracy over visual radiologic assessment. The combined contribution of decreased T1- and increased T2-weighted intensities makes the synthetic FLAIR/T1 contrast particularly effective in MRI-negative HS, setting the basis for broad clinical translation.

CLASSIFICATION OF EVIDENCE

This study provides Class II evidence that in people with TLE and MRI-negative HS, an automated MRI-based classifier accurately determines the side of pathology.

Volumetric glutamate imaging (GluCEST) using 7T MRI can lateralize nonlesional temporal lobe epilepsy: A preliminary study

INTRODUCTION

Drug-resistant epilepsy patients show worse outcomes after resection when standard neuroimaging is nonlesional, which occurs in one-third of patients. In prior work, we employed 2-D glutamate imaging, Glutamate Chemical Exchange Saturation Transfer (GluCEST), to lateralize seizure onset in nonlesional temporal lobe epilepsy (TLE) based on increased ipsilateral GluCEST signal in the total hippocampus and hippocampal head. We present a significant advancement to single-slice GluCEST imaging, allowing for three-dimensional analysis of brain glutamate networks.

METHODS

The study population consisted of four MRI-negative, nonlesional TLE patients (two male, two female) with electrographically identified left temporal onset seizures. Imaging was conducted on a Siemens 7T MRI scanner using the CEST method for glutamate, while the advanced normalization tools (ANTs) pipeline and the Automated Segmentation of the Hippocampal Subfields (ASHS) method were employed for image analysis.

RESULTS

Volumetric GluCEST imaging was validated in four nonlesional TLE patients showing increased glutamate lateralized to the hippocampus of seizure onset (p = .048, with a difference among ipsilateral to contralateral GluCEST signal percentage ranging from -0.05 to 1.37), as well as increased GluCEST signal in the ipsilateral subiculum (p = .034, with a difference among ipsilateral to contralateral GluCEST signal ranging from 0.13 to 1.57).

CONCLUSIONS

The ability of 3-D, volumetric GluCEST to localize seizure onset down to the hippocampal subfield in nonlesional TLE is an improvement upon our previous 2-D, single-slice GluCEST method. Eventually, we hope to expand volumetric GluCEST to whole-brain glutamate imaging, thus enabling noninvasive analysis of glutamate networks in epilepsy and potentially leading to improved clinical outcomes.

SCN2A and Its Related Epileptic Phenotypes

Epilepsies due to SCN2A mutations can present with a broad range of phenotypes that are still not fully understood. Clinical characteristics of SNC2A-related epilepsy may vary from neonatal benign epilepsy to early-onset epileptic encephalopathy, including Ohtahara syndrome and West syndrome, and epileptic encephalopathies occurring at later ages (usually within the first 10 years of life). Some patient may present with intellectual disability and/or autism or movement disorders and without epilepsy. The heterogeneity of the phenotypes associated to such genetic mutations does not always allow the clinician to address his suspect on this gene. For this reason, diagnosis is usually made after a multiple gene panel examination through next generation sequencing (NGS) or after whole exome sequencing (WES) or whole genome sequencing (WGS). Subsequently, confirmation by Sanger sequencing can be obtained. Mutations in SCN2A are inherited as an autosomal dominant trait. Most individuals diagnosed with SCN2A–benign familial neonatal-infantile seizures (BFNIS) have an affected parent; however, hypothetically, a child may present SCN2A-BNFNIS as the result of a de novo pathogenic variant. Almost all individuals with SCN2A and severe epileptic encephalopathies have a de novo pathogenic variant. SNC2A-related epilepsies have not shown a clear genotype–phenotype correlation; in some cases, a same variant may lead to different presentations even within the same family and this could be due to other genetic factors or to environmental causes. There is no “standardized” treatment for SCN2A-related epilepsy, as it varies in relation to the clinical presentation and the phenotype of the patient, according to its own gene mutation. Treatment is based mainly on antiepileptic drugs, which include classic wide-spectrum drugs, such as valproic acid, levetiracetam, and lamotrigine. However, specific agents, which act directly modulating the sodium channels activity (phenytoin, carbamazepine, oxcarbamazepine, lamotrigine, and zonisamide), have shown positive result, as other sodium channel blockers (lidocaine and mexiletine) or even other drugs with different targets (phenobarbital).

Artificial intelligence for brain diseases: A systematic review

Artificial intelligence (AI) is a major branch of computer science that is fruitfully used for analyzing complex medical data and extracting meaningful relationships in datasets, for several clinical aims. Specifically, in the brain care domain, several innovative approaches have achieved remarkable results and open new perspectives in terms of diagnosis, planning, and outcome prediction. In this work, we present an overview of different artificial intelligent techniques used in the brain care domain, along with a review of important clinical applications. A systematic and careful literature search in major databases such as Pubmed, Scopus, and Web of Science was carried out using "artificial intelligence" and "brain" as main keywords. Further references were integrated by cross-referencing from key articles. 155 studies out of 2696 were identified, which actually made use of AI algorithms for different purposes (diagnosis, surgical treatment, intra-operative assistance, and postoperative assessment). Artificial neural networks have risen to prominent positions among the most widely used analytical tools. Classic machine learning approaches such as support vector machine and random forest are still widely used. Task-specific algorithms are designed for solving specific problems. Brain images are one of the most used data types. AI has the possibility to improve clinicians' decision-making ability in neuroscience applications. However, major issues still need to be addressed for a better practical use of AI in the brain. To this aim, it is important to both gather comprehensive data and build explainable AI algorithms.