Ella Sahlas

Ella Sahlas

Auto-generated profile, based on publicly available information

Ella Sahlas

McConnell Brain Imaging Centre (BIC) and Centre of Excellence in Epilepsy at The Neuro (CEEN), Montreal Neurological Institute, McGill University, Montreal, Quebec, Canada

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 June 2025 in Neurology
Verified
Journal Article
Open access
ESElla Sahlas
Ella Sahlas
TATamir Avigdor
Tamir Avigdor
ANAlexander Ngo
Alexander Ngo
JRJessica Royer
Jessica Royer
SLSara Lariviere
Sara Lariviere
BBBoris C. Bernhardt
Boris C. Bernhardt
Epilepsy is increasingly conceptualized as a network disorder, and advancing methods for its diagnosis and treatment requires characterizing both the epileptic generator and related networks. Previous research has highlighted alterations in cortical structure and hemodynamics in epilepsy, but it remains unknown whether these alterations concentrate in areas generating epileptic activity. The aim of this study was to interrogate alterations in microstructure, morphology, and intrinsic local function within and beyond spiking tissue in focal epilepsy, combining multimodal MRI and high-density EEG (HD-EEG). In this cross-sectional study, we recruited patients with focal epilepsy from the Montreal Neurological Institute's epilepsy monitoring unit. Age-matched and sex-matched control participants were recruited from the community. The 3T MRI acquisitions included T1-weighted, diffusion, quantitative T1 relaxometry, and resting-state functional MRI. Open-access processing tools derived cortex-wide maps of morphology and microstructure (cortical thickness, mean diffusivity, and quantitative T1 relaxometry) and intrinsic local function (timescales, connectivity distance, and node strength). Multivariate approaches generated structural and functional alteration scores for each cortical location. Using HD-EEG electrical source imaging, the most prominent spike type was localized and we quantified MRI alterations within spike sources, as well as in proximal and connected networks. We included 25 patients with focal epilepsy (48% female, mean ± SD age = 31.28 ± 9.30 years) and 55 control participants: 30 for imaging feature normalization (50% female, 31.40 ± 8.74 years) and 25 for computing normative connectivity (48% female, 31.04 ± 5.65 years). Regions harboring spike sources showed increased structural MRI alterations compared with the rest of the brain (mean ± SE: 27.98% ± 6.37% vs 17.67% ± 3.32%; t = 2.28; p = 0.036; d = 0.446). Structural compromise extended to all regions with close functional coupling to spike sources, but not to anatomical neighbors. In a modest sample, spiking regions contained more marked alterations in microstructure and morphology than the remaining cortex. There were nevertheless broader network effects, which may relate to a cascading of structural changes to connected cortices. These results underscore the utility of combining quantitative MRI and EEG for characterizing epileptogenic tissue and assessing network effects.

25 April 2025 in bioRxiv
Verified
Preprint
Open access
PDF file
KXKe Xie
Ke Xie
ESElla Sahlas
Ella Sahlas
ANAlexander Ngo
Alexander Ngo
JCJudy Chen
Judy Chen
TAThaera Arafat
Thaera Arafat
BBBoris C. Bernhardt
Boris C. Bernhardt
Temporal lobe epilepsy (TLE) presents with substantial inter-patient variability in clinical and neuroimaging manifestations. This multicenter study examined inter-individual differences in spatial patterns of intrinsic brain function in TLE using normative modeling at multiple spatial scales and evaluated the effectiveness of individual functional deviations for clinical diagnosis and postsurgical outcome prediction. We analyzed multimodal MRI data on 298 healthy controls, 282 TLE patients, and 45 disease controls with extratemporal epilepsy. Cortical function was profiled at local, regional, and global scales using brain signal variability, regional homogeneity, and node strength. We estimated patient-specific W -score maps to index deviations from normative metrics. Compared to healthy controls, patients with TLE showed considerable variations in patterns of functional alterations across the cortex, with the highest overlap in the ipsilateral mesiotemporal regions. Connectome-based simulation revealed the paralimbic and medial default mode regions as key disease epicenters. Functional changes were primarily underpinned by superficial white matter anomalies. Supervised pattern learning achieved classification AUCs of 0.76 for TLE versus disease controls, 0.74 for left versus right TLE, and 0.63 for seizure-free versus non-seizure-free TLE, with greater contralateral temporal functional deviations correlating with unfavorable postsurgical seizure outcome. Our findings reveal the heterogeneous impact of TLE on intrinsic cortical function. These biomarkers hold promise for clinical translation, guiding precision therapeutics and enhancing presurgical decision-making in TLE.

18 May 2025 in bioRxiv
Verified
Preprint
Open access
PDF file
RCRaul Rodriguez Cruces
Raul Rodriguez Cruces
JLJack Lam
Jack Lam
TAThaera Arafat
Thaera Arafat
JRJessica Royer
Jessica Royer
JCJudy Chen
Judy Chen
BBBoris C. Bernhardt
Boris C. Bernhardt
Temporal lobe epilepsy (TLE), the most common pharmaco-resistant epilepsy in adults, has been linked to structural brain changes extending beyond the mesiotemporal areas. While not traditionally viewed as a neurodegenerative disorder, recent ex-vivo studies have shown elevated levels of misfolded tau protein in TLE. This study investigated tau deposition in TLE patients using the in-vivo PET tracer [18F]MK-6240. We studied 28 TLE patients and 28 healthy controls to assess tau uptake and its relationship with brain connectivity, clinical variables, and cognitive function alongside post-surgical tissue from a subset of patients. Compared to controls, TLE patients exhibited markedly increased [18F]MK-6240 uptake in bilateral superior and medial temporal regions and the parietal cortex, with tau accumulation following regional functional and structural connectivity and cognitive impairment. Immunohistochemistry analysis confirmed variable phosphorylated tau staining in 5/6 operated cases with available specimens. These findings suggest that tau accumulation contributes to cognitive decline observed in TLE, supporting a potential role of tau in epilepsy-related neurodegeneration.

20 September 2025 in bioRxiv
Verified
Preprint
Open access
JCJudy Chen
Judy Chen
ESElla Sahlas
Ella Sahlas
YZYigu Zhou
Yigu Zhou
NCNatalie Chen
Natalie Chen
JXJim Xie
Jim Xie
BBBoris C. Bernhardt
Boris C. Bernhardt
Background: The application of artificial intelligence (AI)/machine learning (ML) to MRI can be a powerful tool to streamline clinical decision-making, yet variability amongst MRI sequences and algorithms have hindered appropriate assessment of reliability and generalizability. Methods: We conducted a systematic review and meta-analysis of the ability of current AI/ML models operating on MRI data for: 1) epilepsy diagnosis, 2) temporal lobe epilepsy lateralization, 3) lesion localization, and 4) post-surgical outcome prediction. Searches were conducted across PubMed, Medline, and Embase databases from inception until January 1, 2025. We selected studies that employed AI/ML models trained on any MRI modality to classify at least ten patients across the four main objectives for qualitative assessment, and further included in the meta-analysis if they reported an accuracy rate. The primary outcome of the meta-analysis was the overall accuracy of AI/ML models trained on MRI data. The secondary outcome was the concomitant risk of bias evaluation using PROBAST. Findings: We identified 158 studies for qualitative evaluation and 127 studies for inclusion in the meta-analysis. AI/ML on multimodal MRI could accurately distinguish epilepsy patients from healthy controls (overall accuracy: 88% [85-90]), lateralize temporal lobe epilepsy (90% [87-93]), localize epileptogenic lesions (82% [74-88]), and predict post-surgical seizure-freedom (83% [78-87]). Overall, a high risk of bias remains in the literature; participant bias remained high across all outcomes (64-87%), as well as predictor (88-100%) and analysis (83-100%) bias. Outcome bias was low only for AI/ML studies predicting post-surgical outcomes. Interpretation: Our results support promising accuracy of AI/ML models in epilepsy diagnostics and prognostics but remain highly susceptible to bias in participants, predictors, outcome, and analysis domains which limits current translation to routine clinical practice. We encourage closer interdisciplinary collaboration between clinical and scientific groups to improve validation studies based on thorough study design, analysis, and reporting.

1 December 2025 in bioRxiv
Verified
Preprint
Open access
DCDonna Gift Cabalo
Donna Gift Cabalo
JDJordan DeKraker
Jordan DeKraker
KXKe Xie
Ke Xie
TAThaera Arafat
Thaera Arafat
JRJessica Royer
Jessica Royer
BBBoris C. Bernhardt
Boris C. Bernhardt
Declarative memory deficits represent considerable challenges to adequate functioning and wellbeing in temporal lobe epilepsy (TLE) and frontal lobe epilepsy (FLE), two of the most common pharmaco-resistant epilepsies. TLE and FLE are impacted differently however, with TLE affecting primarily episodic, and mildly semantic memory, while FLE presenting with overall lesser declarative impairments and a greater involvement of language processes. Although functional magnetic resonance imaging (fMRI) studies are overall compatible with differential disruptions of medial temporal and fronto-limbic networks in both syndromes, direct comparisons of brain activity and connectivity remain scarce. The current study investigated how alterations in intrinsic functional brain organization, as mapped with resting-state connectivity (rsFC), shapes altered brain network activations in both episodic and semantic memory states. To this end, we acquired task- and rs-fMRI data in 28 TLE patients and 17 FLE patients and 87 age- and sex-matched healthy controls (HCs). We used activity flow mapping (AFM), a generative machine learning technique that derives plausible task activation patterns via individualized rsFC as well as normative task activation data. Previous studies in HC as well as patient populations have shown that this technique has been effective in identifying mechanisms contributing to atypical functional organization across different context. Overall, AFM reliably predicted task activations across HC, TLE, and FLE, but prediction accuracy was consistently reduced in both patient groups, indicating impaired propagation of task-relevant signals. These reductions co-occurred with abnormal episodic and semantic task activation patterns, atypical rsFC, and behavioral profiles marked by preserved semantic but impaired episodic memory performance in both patient groups. Importantly, although neither task-evoked abnormalities nor rsFC disruptions alone were associated with clinical variables, lower AFM accuracy in patients robustly tracked poorer episodic and semantic memory performance and longer disease duration. Prediction accuracy was driven more by intrinsic functional than structural network features, suggesting that altered network communication, rather than gross anatomy, constrains AFM in pharmaco-resistant epilepsy. Our work revealed syndrome-specific yet convergent disruptions in paralimbic and heteromodal association systems, linking intrinsic dysconnectivity to a shared episodic vulnerability and semantic resilience across TLE and FLE. By modelling how intrinsic connectivity shapes task-evoked responses, AFM provides a mechanistic account of imbalanced memory-state activations and isolates network-flow pathways as targets for intervention and rehabilitation.

22 October 2024 in bioRxiv
Verified
Preprint
Open access
PDF file
ESElla Sahlas
Ella Sahlas
TATamir Avigdor
Tamir Avigdor
ANAlexander Ngo
Alexander Ngo
JRJessica Royer
Jessica Royer
SLSara Larivière
Sara Larivière
BBBoris C. Bernhardt
Boris C. Bernhardt
INTRODUCTION: Epilepsy is increasingly conceptualized as a network disorder, and advancing methods for its diagnosis and treatment requires characterizing both the epileptic generator and related networks. We combined multimodal magnetic resonance imaging (MRI) and high-density electroencephalography (HD-EEG) to interrogate alterations in cortical microstructure, morphology, and intrinsic local function within and beyond spiking tissue in focal epilepsy. METHODS: We studied 25 patients with focal epilepsy (12F, mean ± SD age = 31.28 ± 9.30 years) and 55 age- and sex-matched healthy controls, subdivided into a group of 30 for imaging feature normalization (15F, 31.40 ± 8.74 years) and a group of 25 for replication (12F, 31.04 ± 5.65 years). The 3T MRI acquisition included T1-weighted, diffusion, quantitative T1 relaxometry, and resting-state functional imaging. Open-access MRI processing tools derived cortex-wide maps of morphology and microstructure (cortical thickness, mean diffusivity, and quantitative T1 relaxometry) and intrinsic local function and connectivity (timescales, connectivity distance, and node strength) for all participants. Multivariate approaches generated structural and functional alteration scores for each cortical location. Using HD-EEG electrical source imaging, the most prominent spike type was localized and we quantified MRI alterations within spike sources, as well as in proximal and connected networks. RESULTS: Regions harboring spike sources showed increased structural MRI alterations compared to the rest of the brain in patients. Structural compromise extended to all regions with close functional coupling to spike sources, but not to anatomical neighbors of spike sources. This finding was replicated using average control functional and anatomical matrices instead of patient-specific matrices. CONCLUSION: Spiking regions contain more marked alterations in microstructure and morphology than the remaining cortex, and combining imaging with neurophysiology techniques may ultimately help identify the epileptogenic zone non-invasively. There are nevertheless broader networks effects, which may relate to a cascading of structural changes to functionally connected cortices. These results underscore the utility of combining high-definition MRI and EEG approaches for characterizing epileptogenic tissue and assessing distributed network effects.