Browsing by Author "Klein, Karl Martin"

Now showing 1 - 4 of 4
  • Thumbnail Image
    ItemOpen Access
    Identifying Somatic Variants using DNA Derived from Stereo-Electroencephalography Electrodes in Patients with Focal Epilepsy
    (2024-11-04) Mascarenhas, Rumika; Klein, Karl Martin; Tarailo-Graovac, Maja; Kurrasch, Deborah; Wiebe, Samuel
    Brain somatic variants play a crucial role in the etiology of focal epilepsy. Detecting these variants is challenging due to their presence in a subset of cells, resulting in a reduced variant allele frequency (VAF). Traditional methods rely on brain tissue obtained during resective epilepsy surgery, limiting accessibility and applicability, especially in patients with non-lesional epilepsies who are less likely to undergo surgery. In response to these limitations, a novel approach utilizes DNA extracted from depth electrodes employed in stereo-electroencephalography (SEEG) procedures. This method offers several advantages over resected brain tissue, such as the inclusivity of patients not undergoing surgery and access to multiple brain regions through implanted depth electrodes. Recent studies demonstrated the feasibility of detecting somatic variants using SEEG-derived DNA, highlighting its potential in non-lesional epilepsies. However, challenges remain, including potential cell contaminations and lower cell yields, necessitating DNA amplification that introduces associated artifacts. This thesis introduces an improved SEEG harvesting protocol addressing these issues. Our optimized technique purifies neuronal nuclei, mitigating cell contaminations, and incorporates a newer amplification method to minimize artifacts. Additionally, the thesis introduces the implementation of quality control steps for sample selection and a bioinformatic workflow to reduce artifactual and false positive variants, enhancing the reliability of downstream variant identification. With these improvements, this project aims to enhance the reliability and applicability of SEEG-derived DNA in understanding the molecular basis of focal epilepsy, paving the way for diagnosis and improved treatment strategies.
  • Thumbnail Image
    ItemOpen Access
    Machine learning using multimodal clinical, electroencephalographic, and magnetic resonance imaging data can predict incident depression in adults with epilepsy: A pilot study
    (Wiley, 2023-07-08) Delgado-García, Guillermo; Engbers, Jordan D. T.; Wiebe, Samuel; Mouches, Pauline; Amador, Kimberly; Forkert, Nils D.; White, James; Sajobi, Tolulope; Klein, Karl Martin; Josephson, Colin B.; Calgary Comprehensive Epilepsy Program Collaborators
    Objective: To develop a multi-modal machine learning (ML) approach for predicting incident depression in adults with epilepsy. Methods: We randomly selected 200 patients from the Calgary Comprehensive Epilepsy Program registry and linked their registry-based clinical data to their first-available clinical electroencephalogram (EEG) and magnetic resonance imaging (MRI) study. We excluded patients with a clinical or Neurological Disorders Depression Inventory for Epilepsy (NDDI-E)-based diagnosis of major depression at baseline. The NDDI-E was used to detect incident depression over a median of 2.4 years follow-up (IQR 1.5-3.3 years). A ReliefF algorithm was applied to clinical as well as quantitative EEG and MRI parameters for feature selection. Six ML algorithms were trained and tested using stratified three-fold cross-validation. Multiple metrics were used to assess model performances. Results: Of 200 patients, 150 had EEG and MRI data of sufficient quality for ML, of which 59 were excluded due to prevalent depression. Therefore, 91 patients (41 women) were included with a median age of 29 (IQR 22-44) years. A total of 42 features were selected by ReliefF, none of which were quantitative MRI or EEG variables. All models had a sensitivity >80% and 5 of 6 had an F1 score ≥0.72. Multilayer perceptron model had the highest F1 score (median 0.74; interquartile range [IQR] 0.71-0.78) and sensitivity (84.3%). Median area under the receiver operating characteristic curve and normalized Matthews correlation coefficient were 0.70 (IQR 0.64-0.78) and 0.57 (IQR 0.50-0.65), respectively. Significance: Multimodal machine learning using baseline features can predict incident depression in this population. Our pilot models demonstrated high accuracy for depression prediction. However, overall performance and calibration can be improved. This model has promise for identifying those at risk for incident depression during follow-up, though efforts to refine it in larger populations along with external validation are required.
  • Thumbnail Image
    ItemOpen Access
    Predicting the Side Effects of Antiseizure Medications Using Machine Learning Models
    (2024-01-02) Lin, Chantelle Qing Yang; Josephson, Colin Bruce; Sajobi, Tolulope; Klein, Karl Martin; Forkert, Nils Daniel
    With over 20 anti-seizure medications (ASMs), identifying the ideal drug is often imprecise and time-consuming. Developing predictive models to expedite optimal drug selection is challenging due to the minimal differences in efficacy among adult patients with epilepsy. However, side-effects vary considerably between medications, and are one of the main reasons for discontinuation of ASM treatment. The aim was to (1) assess the prognostic utility of high- dimensional data such as genetic features with clinical features to predict ASM discontinuation, and (2) determine the optimal regression/machine learning model for predicting ASM discontinuation. This retrospective cohort study included 4,853 exposures to any ASM, and 624 patients exposed to valproic acid (VPA) from the RAISE-GENIC study during the years 2006-2020. The predicted outcome was defined as ASM discontinuation due to any side-effect reported by the patient. Clinical features included age of onset, patient age, sex, comorbidities, seizure type, EEG variables, and imaging variables. Network analysis of mRNA expression data from VPA-exposed neurons derived from control induced pluripotent stem cells (iPSCs) was leveraged to extract exome sequencing and genome-wide single nucleotide polymorphism data. Features were selected for model inclusion based on relevance as determined by the ReliefF algorithm. Penalized logistic regression, support vector machine, random forest, and k-nearest neighbor models were trained on the normalized bootstrapped dataset and model quality was assessed using stratified 10-fold cross validation. Models with only clinical and combined clinical and genetic features were compared by quantitative as well as visual discrimination and calibration metrics. The results showed that the best performing model was the penalized logistic regression using the VPA dataset with genetic and clinical features. The accuracy was 0.75 [95% confidence interval 0.74-0.76], area under the receiver operating characteristic curve was 0.66 [0.66-0.67], Brier score was 0.20 [0.19-0.21], sensitivity was 0.42 [0.41-0.42], and specificity 0.82 [0.82-0.83]. Machine learning using clinical and genetic features can moderately predict treatment-ending side- effects to VPA with moderate performance, discrimination, and calibration. If these results can be validated and improved upon, decision tools can be incorporated into clinical routines, simplifying drug prescriptions, saving time, and improving patient quality of life.
  • Thumbnail Image
    ItemOpen Access
    The negative BOLD response as a marker of the seizure onset zone
    (2023-09-20) Dykens, Perry Everett; Federico, Paolo; Goodyear, Brad; Klein, Karl Martin
    Epilepsy is a neurological disease affecting 70 million people worldwide. For most individuals, these seizures can be controlled using medications, however nearly 1 in 3 people may need surgery to achieve seizure freedom. For this surgery to be successful, the brain region generating the seizures, which contains the critical seizure onset zone (SOZ), must be accurately identified and removed. Unfortunately, the surgical success rate is low likely due to imprecise determination of the SOZ. As a novel approach to SOZ identification, the collection of intracranial electroencephalography and functional magnetic resonance imaging (iEEG-fMRI) has been proposed as a novel method of identifying the SOZ. However, iEEG-fMRI faces the methodological challenge of artifact introduced from MR scanning which completely obscures the physiological EEG signal. Therefore, the first step towards bringing iEEG-fMRI into the clinical realm is to improve methods for extracting the physiological EEG signal from the iEEG-fMRI data. To this end, the first study in this thesis validated a set of methods aimed at removing fMRI artifact from iEEG, culminating in the creation of the first automatic iEEG pre-processing pipeline. The next step towards clinical utility for iEEG-fMRI is improving our interpretation of iEEG-fMRI results. Traditionally, only positive IED related fMRI activation maps were considered in relation to SOZ localization, and the negative response was ignored. It has been suggested that both positive and negative activation maps should be considered, and the maximal cluster of these two maps, regardless of polarity, should be used to localize the SOZ. In the second study, the concept was tested using iEEG-fMRI and it was found that the use of the maximal negative cluster had limited utility for SOZ localization. The results of this thesis provide a new method for preparing EEG data from iEEG-fMRI experiments and it shows that the bulk of maximal negative fMRI clusters have limited reliability for clinical applications.