Medicine (Austin & Northern Health) - Research Publications
Permanent URI for this collection
Search Results
1 - 10 of 12
-
ItemNo Preview AvailableAutomated Interictal Epileptiform Discharge Detection from Scalp EEG Using Scalable Time-series Classification Approaches(WORLD SCIENTIFIC PUBL CO PTE LTD, 2023-01)Deep learning for automated interictal epileptiform discharge (IED) detection has been topical with many published papers in recent years. All existing works viewed EEG signals as time-series and developed specific models for IED classification; however, general time-series classification (TSC) methods were not considered. Moreover, none of these methods were evaluated on any public datasets, making direct comparisons challenging. This paper explored two state-of-the-art convolutional-based TSC algorithms, InceptionTime and Minirocket, on IED detection. We fine-tuned and cross-evaluated them on a public (Temple University Events - TUEV) and two private datasets and provided ready metrics for benchmarking future work. We observed that the optimal parameters correlated with the clinical duration of an IED and achieved the best area under precision-recall curve (AUPRC) of 0.98 and F1 of 0.80 on the private datasets, respectively. The AUPRC and F1 on the TUEV dataset were 0.99 and 0.97, respectively. While algorithms trained on the private sets maintained their performance when tested on the TUEV data, those trained on TUEV could not generalize well to the private data. These results emerge from differences in the class distributions across datasets and indicate a need for public datasets with a better diversity of IED waveforms, background activities and artifacts to facilitate standardization and benchmarking of algorithms.
-
ItemNo Preview AvailableDeep learning for automated epileptiform discharge detection from scalp EEG: A systematic review(IOP Publishing Ltd, 2022-10-01)Automated interictal epileptiform discharge (IED) detection has been widely studied, with machine learning methods at the forefront in recent years. As computational resources become more accessible, researchers have applied deep learning (DL) to IED detection with promising results. This systematic review aims to provide an overview of the current DL approaches to automated IED detection from scalp electroencephalography (EEG) and establish recommendations for the clinical research community. We conduct a systematic review according to the PRISMA guidelines. We searched for studies published between 2012 and 2022 implementing DL for automating IED detection from scalp EEG in major medical and engineering databases. We highlight trends and formulate recommendations for the research community by analyzing various aspects: data properties, preprocessing methods, DL architectures, evaluation metrics and results, and reproducibility. The search yielded 66 studies, and 23 met our inclusion criteria. There were two main DL networks, convolutional neural networks in 14 studies and long short-term memory networks in three studies. A hybrid approach combining a hidden Markov model with an autoencoder was employed in one study. Graph convolutional network was seen in one study, which considered a montage as a graph. All DL models involved supervised learning. The median number of layers was 9 (IQR: 5-21). The median number of IEDs was 11 631 (IQR: 2663-16 402). Only six studies acquired data from multiple clinical centers. AUC was the most reported metric (median: 0.94; IQR: 0.94-0.96). The application of DL to IED detection is still limited and lacks standardization in data collection, multi-center testing, and reporting of clinically relevant metrics (i.e. F1, AUCPR, and false-positive/minute). However, the performance is promising, suggesting that DL might be a helpful approach. Further testing on multiple datasets from different clinical centers is required to confirm the generalizability of these methods.
-
ItemNo Preview AvailableDiagnostic delay in focal epilepsy: Association with brain pathology and age(W B SAUNDERS CO LTD, 2022-03)PURPOSE: Between 16-77% of patients with newly diagnosed epilepsy report seizures before diagnosis but little is known about the risk factors for diagnostic delay. Here, we examined the association between prior seizures and neuroimaging findings in newly diagnosed focal epilepsy. METHODS: Adults diagnosed with focal epilepsy at First Seizure Clinics (FSC) at the Royal Melbourne Hospital or Austin Health, Melbourne, Australia, between 2000 and 2010 were included. Medical records were audited for seizure history accrued from the detailed FSC interview. Potentially epileptogenic brain abnormality type, location and extent was determined from neuroimaging. Statistical analysis comprised multivariate logistic regression. RESULTS: Of 735 patients, 44% reported seizure/s before the index seizure. Among the 260 individuals with a potentially epileptogenic brain imaging abnormality, 34% reported prior seizures. Of 475 individuals with no abnormality, 50% reported prior seizures (p < 0.001). Patients with post-stroke changes had lower odds of prior seizures (n = 24/95, OR 0.5, p = 0.005) compared to patients without abnormalities, as did patients with high-grade tumors (n = 1/10, OR 0.1, p = 0.04). Abnormality location or extent was not associated with seizures. Prior seizures were inversely associated with age, patients aged >50 years had lower odds compared to those 18-30 years (OR 0.5, p = 0.01). CONCLUSIONS: A history of prior seizures is less common in patients with newly diagnosed focal epilepsy associated with antecedent stroke or high-grade tumor than in those without a lesion, and is also less common in older individuals. These findings may be related to age, biological mechanisms or aspects of diagnosis and assessment of these events.
-
ItemAdjunctive Transdermal Cannabidiol for Adults With Focal Epilepsy A Randomized Clinical Trial(AMER MEDICAL ASSOC, 2022-07-08)IMPORTANCE: Cannabidiol has shown efficacy in randomized clinical trials for drug-resistant epilepsy in specific syndromes that predominantly affect children. However, high-level evidence for the efficacy and safety of cannabidiol in the most common form of drug-resistant epilepsy in adults, focal epilepsy, is lacking. OBJECTIVE: To investigate the efficacy, safety, and tolerability of transdermally administered cannabidiol in adults with drug-resistant focal epilepsy. DESIGN, SETTING, AND PARTICIPANTS: A randomized, double-blind, placebo-controlled, multicenter clinical trial at 14 epilepsy trial centers in Australia and New Zealand. Participants were adults with drug-resistant focal epilepsy receiving a stable regimen of up to 3 antiseizure medications. Data were analyzed from July 2017 to November 2018. INTERVENTIONS: Eligible participants were randomized (1:1:1) to 195-mg or 390-mg transdermal cannabidiol or placebo twice daily for 12 weeks, after which they could enroll in an open-label extension study for up to 2 years. MAIN OUTCOMES AND MEASURES: Seizure frequency was self-reported using a daily diary. The primary efficacy end point was the least squares mean difference in the log-transformed total seizure frequency per 28-day period, adjusted to a common baseline log seizure rate, during the 12-week treatment period. RESULTS: A total of 188 patients (45% male [85 patients] and 54.8% female [103 patients]) with a mean (SD) age of 39.2 (12.78) years were randomized, treated, and analyzed (195-mg cannabidiol, 63 participants; 390-mg cannabidiol, 62 participants; placebo, 63 participants). At week 12 of the double-blind period, there was no difference in seizure frequency between placebo (mean [SD] 2.49 [1.31] seizures per 28 days) and 195-mg cannabidiol (mean [SD] 2.51 [1.15] seizures per 28 days; least squares mean difference, 0.014; 95% CI, -0.175 to 0.203; P = .89) or 390-mg cannabidiol (mean [SD] 2.59 [1.12] seizures per 28 days; least squares mean difference, 0.096; 95% CI, -0.093 to 0.285; P = .32). By month 6 of the open-label extension, 115 patients (60.8%) achieved a seizure reduction of at least 50%. Treatment-emergent adverse events occurred in 50.4% (63 of 125 participants) of the cannabidiol group vs 41.3% (26 of 63 participants) in the placebo group, with a treatment difference of 9.1% (95% CI, -6.0% to 23.6%), and occurred at similar rates in the cannabidiol groups. Few participants discontinued (7% [14 of 188 participants]), and most (98% [171 of 174 participants]) continued into the open-label extension. CONCLUSIONS AND RELEVANCE: Both doses of transdermal cannabidiol were well tolerated and safe. No significant difference in efficacy was observed between cannabidiol and placebo during the double-blind treatment period. The open-label extension demonstrated the long-term safety, tolerability, and acceptability of transdermal cannabidiol delivery. TRIAL REGISTRATION: ACTRN12616000510448 (double-blind); ACTRN12616001455459 (open-label).
-
ItemSub-genic intolerance, ClinVar, and the epilepsies: A whole-exome sequencing study of 29,165 individuals(CELL PRESS, 2021-06-03)Both mild and severe epilepsies are influenced by variants in the same genes, yet an explanation for the resulting phenotypic variation is unknown. As part of the ongoing Epi25 Collaboration, we performed a whole-exome sequencing analysis of 13,487 epilepsy-affected individuals and 15,678 control individuals. While prior Epi25 studies focused on gene-based collapsing analyses, we asked how the pattern of variation within genes differs by epilepsy type. Specifically, we compared the genetic architectures of severe developmental and epileptic encephalopathies (DEEs) and two generally less severe epilepsies, genetic generalized epilepsy and non-acquired focal epilepsy (NAFE). Our gene-based rare variant collapsing analysis used geographic ancestry-based clustering that included broader ancestries than previously possible and revealed novel associations. Using the missense intolerance ratio (MTR), we found that variants in DEE-affected individuals are in significantly more intolerant genic sub-regions than those in NAFE-affected individuals. Only previously reported pathogenic variants absent in available genomic datasets showed a significant burden in epilepsy-affected individuals compared with control individuals, and the ultra-rare pathogenic variants associated with DEE were located in more intolerant genic sub-regions than variants associated with non-DEE epilepsies. MTR filtering improved the yield of ultra-rare pathogenic variants in affected individuals compared with control individuals. Finally, analysis of variants in genes without a disease association revealed a significant burden of loss-of-function variants in the genes most intolerant to such variation, indicating additional epilepsy-risk genes yet to be discovered. Taken together, our study suggests that genic and sub-genic intolerance are critical characteristics for interpreting the effects of variation in genes that influence epilepsy.
-
ItemMachine learning approaches for imaging-based prognostication of the outcome of surgery for mesial temporal lobe epilepsy(WILEY, 2022-05)OBJECTIVES: Around 30% of patients undergoing surgical resection for drug-resistant mesial temporal lobe epilepsy (MTLE) do not obtain seizure freedom. Success of anterior temporal lobe resection (ATLR) critically depends on the careful selection of surgical candidates, aiming at optimizing seizure freedom while minimizing postoperative morbidity. Structural MRI and FDG-PET neuroimaging are routinely used in presurgical assessment and guide the decision to proceed to surgery. In this study, we evaluate the potential of machine learning techniques applied to standard presurgical MRI and PET imaging features to provide enhanced prognostic value relative to current practice. METHODS: Eighty two patients with drug resistant MTLE were scanned with FDG-PET pre-surgery and T1-weighted MRI pre- and postsurgery. From these images the following features of interest were derived: volume of temporal lobe (TL) hypometabolism, % of extratemporal hypometabolism, presence of contralateral TL hypometabolism, presence of hippocampal sclerosis, laterality of seizure onset volume of tissue resected and % of temporal lobe hypometabolism resected. These measures were used as predictor variables in logistic regression, support vector machines, random forests and artificial neural networks. RESULTS: In the study cohort, 24 of 82 (28.3%) who underwent an ATLR for drug-resistant MTLE did not achieve Engel Class I (i.e., free of disabling seizures) outcome at a minimum of 2 years of postoperative follow-up. We found that machine learning approaches were able to predict up to 73% of the 24 ATLR surgical patients who did not achieve a Class I outcome, at the expense of incorrect prediction for up to 31% of patients who did achieve a Class I outcome. Overall accuracies ranged from 70% to 80%, with an area under the receiver operating characteristic curve (AUC) of .75-.81. We additionally found that information regarding overall extent of both total and significantly hypometabolic tissue resected was crucial to predictive performance, with AUC dropping to .59-.62 using presurgical information alone. Incorporating the laterality of seizure onset and the choice of machine learning algorithm did not significantly change predictive performance. SIGNIFICANCE: Collectively, these results indicate that "acceptable" to "good" patient-specific prognostication for drug-resistant MTLE surgery is feasible with machine learning approaches utilizing commonly collected imaging modalities, but that information on the surgical resection region is critical for optimal prognostication.
-
ItemInterictal EEG and ECG for SUDEP Risk Assessment: A Retrospective Multicenter Cohort Study(FRONTIERS MEDIA SA, 2022-03-18)OBJECTIVE: Sudden unexpected death in epilepsy (SUDEP) is the leading cause of epilepsy-related mortality. Although lots of effort has been made in identifying clinical risk factors for SUDEP in the literature, there are few validated methods to predict individual SUDEP risk. Prolonged postictal EEG suppression (PGES) is a potential SUDEP biomarker, but its occurrence is infrequent and requires epilepsy monitoring unit admission. We use machine learning methods to examine SUDEP risk using interictal EEG and ECG recordings from SUDEP cases and matched living epilepsy controls. METHODS: This multicenter, retrospective, cohort study examined interictal EEG and ECG recordings from 30 SUDEP cases and 58 age-matched living epilepsy patient controls. We trained machine learning models with interictal EEG and ECG features to predict the retrospective SUDEP risk for each patient. We assessed cross-validated classification accuracy and the area under the receiver operating characteristic (AUC) curve. RESULTS: The logistic regression (LR) classifier produced the overall best performance, outperforming the support vector machine (SVM), random forest (RF), and convolutional neural network (CNN). Among the 30 patients with SUDEP [14 females; mean age (SD), 31 (8.47) years] and 58 living epilepsy controls [26 females (43%); mean age (SD) 31 (8.5) years], the LR model achieved the median AUC of 0.77 [interquartile range (IQR), 0.73-0.80] in five-fold cross-validation using interictal alpha and low gamma power ratio of the EEG and heart rate variability (HRV) features extracted from the ECG. The LR model achieved the mean AUC of 0.79 in leave-one-center-out prediction. CONCLUSIONS: Our results support that machine learning-driven models may quantify SUDEP risk for epilepsy patients, future refinements in our model may help predict individualized SUDEP risk and help clinicians correlate predictive scores with the clinical data. Low-cost and noninvasive interictal biomarkers of SUDEP risk may help clinicians to identify high-risk patients and initiate preventive strategies.
-
ItemAssociation Between Psychiatric Comorbidities and Mortality in Epilepsy(LIPPINCOTT WILLIAMS & WILKINS, 2021-10)OBJECTIVE: To explore the impact of psychiatric comorbidities on all-cause mortality in adults with epilepsy from a cohort of patients admitted for video-EEG monitoring (VEM) over 2 decades. METHODS: A retrospective medical record audit was conducted on 2,709 adults admitted for VEM and diagnosed with epilepsy at 3 Victorian comprehensive epilepsy programs from 1995 to 2015. A total of 1,805 patients were identified in whom the record of a clinical evaluation by a neuropsychiatrist was available, excluding 27 patients who died of a malignant brain tumor known at the time of VEM admission. Epilepsy and lifetime psychiatric diagnoses were determined from consensus opinion of epileptologists and neuropsychiatrists involved in the care of each patient. Mortality and cause of death were determined by linkage to the Australian National Death Index and National Coronial Information System. RESULTS: Compared with the general population, mortality was higher in people with epilepsy (PWE) with a psychiatric illness (standardized mortality ratio [SMR] 3.6) and without a psychiatric illness (SMR 2.5). PWE with a psychiatric illness had greater mortality compared with PWE without (hazard ratio 1.41, 95% confidence interval 1.02-1.97) after adjusting for age and sex. No single psychiatric disorder by itself conferred increased mortality in PWE. The distribution of causes of death remained similar between PWE with psychiatric comorbidities and those without. CONCLUSION: The presence of comorbid psychiatric disorders in adults with epilepsy is associated with increased mortality, highlighting the importance of identifying and treating psychiatric comorbidities in these patients.
-
ItemInflammation, ictogenesis, and epileptogenesis: An exploration through human disease(WILEY, 2021-02)Epilepsy is seen historically as a disease of aberrant neuronal signaling manifesting as seizures. With the discovery of numerous auto-antibodies and the subsequent growth in understanding of autoimmune encephalitis, there has been an increasing emphasis on the contribution of the innate and adaptive immune system to ictogenesis and epileptogenesis. Pathogenic antibodies, complement activation, CD8+ cytotoxic T cells, and microglial activation are seen, to various degrees, in different seizure-associated neuroinflammatory and autoimmune conditions. These aberrant immune responses are thought to cause disruptions in neuronal signaling, generation of acute symptomatic seizures, and, in some cases, the development of long-term autoimmune epilepsy. Although early treatment with immunomodulatory therapies improves outcomes in autoimmune encephalitides and autoimmune epilepsies, patient identification and treatment selection are not always clear-cut. This review examines the role of the different components of the immune system in various forms of seizure disorders including autoimmune encephalitis, autoimmune epilepsy, Rasmussen encephalitis, febrile infection-related epilepsy syndrome (FIRES), and new-onset refractory status epilepticus (NORSE). In particular, the pathophysiology and unique cytokine profiles seen in these disorders and their links with diagnosis, prognosis, and treatment decision-making are discussed.
-
ItemAntiepileptic Drug Teratogenicity and De Novo Genetic Variation Load(WILEY, 2020-06)OBJECTIVE: The mechanisms by which antiepileptic drugs (AEDs) cause birth defects (BDs) are unknown. Data suggest that AED-induced BDs may result from a genome-wide increase of de novo variants in the embryo, a mechanism that we investigated. METHODS: Whole exome sequencing data from child-parent trios were interrogated for de novo single-nucleotide variants/indels (dnSNVs/indels) and de novo copy number variants (dnCNVs). Generalized linear models were applied to assess de novo variant burdens in children exposed prenatally to AEDs (AED-exposed children) versus children without BDs not exposed prenatally to AEDs (AED-unexposed unaffected children), and AED-exposed children with BDs versus those without BDs, adjusting for confounders. Fisher exact test was used to compare categorical data. RESULTS: Sixty-seven child-parent trios were included: 10 with AED-exposed children with BDs, 46 with AED-exposed unaffected children, and 11 with AED-unexposed unaffected children. The dnSNV/indel burden did not differ between AED-exposed children and AED-unexposed unaffected children (median dnSNV/indel number/child [range] = 3 [0-7] vs 3 [1-5], p = 0.50). Among AED-exposed children, there were no significant differences between those with BDs and those unaffected. Likely deleterious dnSNVs/indels were detected in 9 of 67 (13%) children, none of whom had BDs. The proportion of cases harboring likely deleterious dnSNVs/indels did not differ significantly between AED-unexposed and AED-exposed children. The dnCNV burden was not associated with AED exposure or birth outcome. INTERPRETATION: Our study indicates that prenatal AED exposure does not increase the burden of de novo variants, and that this mechanism is not a major contributor to AED-induced BDs. These results can be incorporated in routine patient counseling. ANN NEUROL 2020;87:897-906.