arXiv subjects

Yujiang Wang

Publications and source records attributed to Yujiang Wang.

At least 55 records · Page 3Linked to original sources

arXiv2024

Alpha rhythm slowing in temporal epilepsy across Scalp EEG and MEG

EEG slowing is reported in various neurological disorders including Alzheimer's, Parkinson's and Epilepsy. Here, we investigate alpha rhythm slowing in individuals with refractory temporal lobe epilepsy (TLE), compared to healthy controls, using scalp electroencephalography (EEG) and magnetoencephalography (MEG). We retrospectively analysed data from 17,(46) healthy controls and 22,(24) individuals with TLE who underwent scalp EEG and (MEG) recordings as part of presurgical evaluation. Resting-state, eyes-closed recordings were source reconstructed using the standardized low-resolution brain electrographic tomography (sLORETA) method. We extracted low (slow) 6-9 Hz and high (fast) 10-11 Hz alpha relative band power and calculated the alpha power ratio by dividing low (slow) alpha by high (fast) alpha. This ratio was computed for all brain regions in all individuals. Alpha oscillations were slower in individuals with TLE than controls (p<0.05). This effect was present in both the ipsilateral and contralateral hemispheres, and across widespread brain regions. Alpha slowing in TLE was found in both EEG and MEG recordings. We interpret greater low (slow)-alpha as greater deviation from health.

q-bio.NC↗

arXiv2024

Manifold-valued models for analysis of EEG time series data

We propose a model for time series taking values on a Riemannian manifold and fit it to time series of covariance matrices derived from EEG data for patients suffering from epilepsy. The aim of the study is two-fold: to develop a model with interpretable parameters for different possible modes of EEG dynamics, and to explore the extent to which modelling results are affected by the choice of manifold and its associated geometry. The model specifies a distribution for the tangent direction vector at any time point, combining an autoregressive term, a mean reverting term and a form of Gaussian noise. Parameter inference is carried out by maximum likelihood estimation, and we compare modelling results obtained using the standard Euclidean geometry on covariance matrices and the affine invariant geometry. Results distinguish between epileptic seizures and interictal periods between seizures in patients: between seizures the dynamics have a strong mean reverting component and the autoregressive component is missing, while for the majority of seizures there is a significant autoregressive component and the mean reverting effect is weak. The fitted models are also used to compare seizures within and between patients. The affine invariant geometry is advantageous and it provides a better fit to the data.

stat.ME↗

arXiv2024

Medical records condensation: a roadmap towards healthcare data democratisation

The prevalence of artificial intelligence (AI) has envisioned an era of healthcare democratisation that promises every stakeholder a new and better way of life. However, the advancement of clinical AI research is significantly hurdled by the dearth of data democratisation in healthcare. To truly democratise data for AI studies, challenges are two-fold: 1. the sensitive information in clinical data should be anonymised appropriately, and 2. AI-oriented clinical knowledge should flow freely across organisations. This paper considers a recent deep-learning advent, dataset condensation (DC), as a stone that kills two birds in democratising healthcare data. The condensed data after DC, which can be viewed as statistical metadata, abstracts original clinical records and irreversibly conceals sensitive information at individual levels; nevertheless, it still preserves adequate knowledge for learning deep neural networks (DNNs). More favourably, the compressed volumes and the accelerated model learnings of condensed data portray a more efficient clinical knowledge sharing and flowing system, as necessitated by data democratisation. We underline DC's prospects for democratising clinical data, specifically electrical healthcare records (EHRs), for AI research through experimental results and analysis across three healthcare datasets of varying data types.

cs.LG↗

arXiv2023

Can Large Language Models Be Good Companions? An LLM-Based Eyewear System with Conversational Common Ground

Developing chatbots as personal companions has long been a goal of artificial intelligence researchers. Recent advances in Large Language Models (LLMs) have delivered a practical solution for endowing chatbots with anthropomorphic language capabilities. However, it takes more than LLMs to enable chatbots that can act as companions. Humans use their understanding of individual personalities to drive conversations. Chatbots also require this capability to enable human-like companionship. They should act based on personalized, real-time, and time-evolving knowledge of their owner. We define such essential knowledge as the \textit{common ground} between chatbots and their owners, and we propose to build a common-ground-aware dialogue system from an LLM-based module, named \textit{OS-1}, to enable chatbot companionship. Hosted by eyewear, OS-1 can sense the visual and audio signals the user receives and extract real-time contextual semantics. Those semantics are categorized and recorded to formulate historical contexts from which the user's profile is distilled and evolves over time, i.e., OS-1 gradually learns about its user. OS-1 combines knowledge from real-time semantics, historical contexts, and user-specific profiles to produce a common-ground-aware prompt input into the LLM module. The LLM's output is converted to audio, spoken to the wearer when appropriate.We conduct laboratory and in-field studies to assess OS-1's ability to build common ground between the chatbot and its user. The technical feasibility and capabilities of the system are also evaluated. OS-1, with its common-ground awareness, can significantly improve user satisfaction and potentially lead to downstream tasks such as personal emotional support and assistance.

cs.HC↗

arXiv2023

Incomplete resection of the icEEG seizure onset zone is not associated with post-surgical outcomes

Delineation of seizure onset regions from EEG is important for effective surgical workup. However, it is unknown if their complete resection is required for seizure freedom, or in other words, if post-surgical seizure recurrence is due to incomplete removal of the seizure onset regions. Retrospective analysis of icEEG recordings from 63 subjects (735 seizures) identified seizure onset regions through visual inspection and algorithmic delineation. We analysed resection of onset regions and correlated this with post-surgical seizure control. Most subjects had over half of onset regions resected (70.7% and 60.5% of subjects for visual and algorithmic methods, respectively). In investigating spatial extent of onset or resection, and presence of diffuse onsets, we found no substantial evidence of association with post-surgical seizure control (all AUC<0.7, p>0.05). Seizure onset regions tends to be at least partially resected, however a less complete resection is not associated with worse post-surgical outcome. We conclude that seizure recurrence after epilepsy surgery is not necessarily a result of failing to completely resect the seizure onset zone, as defined by icEEG. Other network mechanisms must be involved, which are not limited to seizure onset regions alone.

q-bio.NC↗

arXiv2023

Multivariate brain-cognition associations in euthymic bipolar disorder

Background: People with bipolar disorder (BD) tend to show widespread cognitive impairment compared to healthy controls. Impairments in processing speed (PS), attention, and executive function (EF) may represent 'core' impairments that have a role in wider cognitive dysfunction. Cognitive impairments appear to relate to structural brain abnormalities in BD, but whether core deficits are related to particular brain regions is unclear and much of the research on brain-cognition associations is limited by univariate analysis and small samples. Methods: Euthymic BD patients (n=56) and matched healthy controls (n=26) underwent T1-weighted MRI scans and completed neuropsychological tests of PS, attention, and EF. We utilised public datasets to develop a normative model of cortical thickness (n=5,977) to generate robust estimations of cortical abnormalities in patients. Canonical correlation analysis was used to assess multivariate brain-cognition associations in BD, controlling for age, sex, and premorbid IQ. Results: BD showed impairments on tests of PS, attention, and EF, and abnormal cortical thickness in several brain regions compared to healthy controls. Impairments in tests of PS and EF were most strongly associated with cortical thickness in left inferior temporal, right entorhinal, and right temporal pole areas. Conclusion: Impairments in PS, attention, and EF can be observed in euthymic BD and may be related to abnormal cortical thickness in temporal regions. Future research should continue to leverage multivariate methods to examine complex brain-cognition associations in BD. Future research may benefit from exploring covariance between traditional brain structural morphological metrics such as cortical thickness, cortical volume, and surface area.

q-bio.NC↗

arXiv2023

Complementary structural and functional abnormalities to localise epileptogenic tissue

When investigating suitability for surgery, people with drug-refractory focal epilepsy may have intracranial EEG (iEEG) electrodes implanted to localise seizure onset. Diffusion-weighted magnetic resonance imaging (dMRI) may be acquired to identify key white matter tracts for surgical avoidance. Here, we investigate whether structural connectivity abnormalities, inferred from dMRI, may be used in conjunction with functional iEEG abnormalities to aid localisation and resection of the epileptogenic zone (EZ), and improve surgical outcomes in epilepsy. We retrospectively investigated data from 43 patients with epilepsy who had surgery following iEEG. Twenty five patients (58%) were free from disabling seizures (ILAE 1 or 2) at one year. For all patients, T1-weighted and diffusion-weighted MRIs were acquired prior to iEEG implantation. Interictal iEEG functional, and dMRI structural connectivity abnormalities were quantified by comparison to a normative map and healthy controls respectively. First, we explored whether the resection of maximal (dMRI and iEEG) abnormalities related to improved surgical outcomes. Second, we investigated whether the modalities provided complementary information for improved prediction of surgical outcome. Third, we suggest how dMRI abnormalities may be useful to inform the placement of iEEG electrodes as part of the pre-surgical evaluation using a patient case study. Seizure freedom was 15 times more likely in those patients with resection of maximal dMRI and iEEG abnormalities (p=0.008). Both modalities were separately able to distinguish patient outcome groups and when combined, a decision tree correctly separated 36 out of 43 (84%) patients based on surgical outcome. Structural dMRI could be used in pre-surgical evaluations, particularly when localisation of the EZ is uncertain, to inform personalised iEEG implantation and resection.

q-bio.NC↗

arXiv2023

Aperiodic MEG abnormality in patients with focal to bilateral tonic-clonic seizures

Aperiodic activity is a physiologically distinct component of the electrophysiological power spectrum. It is suggested to reflect the balance of excitation and inhibition in the brain, within selected frequency bands. However, the impact of recurrent seizures on aperiodic activity remains unknown, particularly in patients with severe bilateral seizures. Here, we hypothesised greater aperiodic abnormality in the epileptogenic zone, in patients with focal to bilateral tonic clonic (FBTC) seizures, and earlier age of seizure onset. Pre-operative magnetoencephalography (MEG) recordings were acquired from 36 patients who achieved complete seizure freedom (Engel I outcome) post-surgical resection. A normative whole brain map of the aperiodic exponent was computed by averaging across subjects for each region in the hemisphere contralateral to the side of resection. Selected regions of interest were then tested for abnormality using deviations from the normative map in terms of z-scores. Resection masks drawn from postoperative structural imaging were used as an approximation of the epileptogenic zone. Patients with FBTC seizures had greater abnormality compared to patients with focal onset seizures alone in the resection volume (p=0.003, area under the ROC curve = 0.78 ). Earlier age of seizure onset was correlated with greater abnormality of the aperiodic exponent in the resection volume (correlation coefficient = -0.3, p= 0.04)) as well as the whole cortex (rho = -0.33, p=0.03). The abnormality of the aperiodic exponent did not significantly differ between the resected and non-resected regions of the brain. Abnormalities in aperiodic components relate to important clinical characteristics such as severity and age of seizure onset. This suggests the potential use of the aperiodic band power component as a marker for severity of epilepsy.

q-bio.NC↗

arXiv2023

Identifying epileptogenic abnormality by decomposing intracranial EEG and MEG power spectra

Identifying abnormal electroencephalographic activity is crucial in diagnosis and treatment of epilepsy. Recent studies showed that decomposing brain activity into periodic (oscillatory) and aperiodic (trend across all frequencies) components may illuminate drivers of changes in spectral activity. Using iEEG data from 234 subjects, we constructed a normative map and compared this with a separate cohort of 63 patients with refractory focal epilepsy being considered for neurosurgery. The normative map was computed using three approaches: (i) relative complete band power, (ii) relative band power with the aperiodic component removed (iii) the aperiodic exponent. Corresponding abnormalities were also calculated for each approach in the separate patient cohort. We investigated the spatial profiles of the three approaches, assessed their localizing ability, and replicated our findings in a separate modality using MEG. The normative maps of relative complete band power and relative periodic band power had similar spatial profiles. In the aperiodic normative map, exponent values were highest in the temporal lobe. Abnormality estimated through the complete band power robustly distinguished between good and bad outcome patients. Neither periodic band power nor aperiodic exponent abnormalities distinguished seizure outcome groups. Combining periodic and aperiodic abnormalities improved performance, similar to the complete band power approach. Our findings suggest that sparing cerebral tissue that generates abnormalities in either periodic or aperiodic activity may lead to a poor surgical outcome. Both periodic and aperiodic abnormalities are necessary to distinguish patient outcomes, with neither sufficient in isolation. Future studies could investigate whether periodic or aperiodic abnormalities are affected by the cerebral location or pathology.

q-bio.NC↗

arXiv2023

Chronic iEEG recordings and interictal spike rate reveal multiscale temporal modulations in seizure states

Background and Objectives: Many biological processes are modulated by rhythms on circadian and multidien timescales. In focal epilepsy, various seizure features, such as spread and duration, can change from one seizure to the next within the same patient. However, the specific timescales of this variability, as well as the specific seizure characteristics that change over time, are unclear. Methods: Here, in a cross-sectional observational study, we analysed within-patient seizure variability in 10 patients with chronic intracranial EEG recordings (185-767 days of recording time, 57-452 analysed seizures/patient). We characterised the seizure evolutions as sequences of a finite number of patient-specific functional seizure network states (SNSs). We then compared SNS occurrence and duration to (1) time since implantation and (2) patient-specific circadian and multidien cycles in interictal spike rate. Results: In most patients, the occurrence or duration of at least one SNS was associated with the time since implantation. Some patients had one or more SNSs that were associated with phases of circadian and/or multidien spike rate cycles. A given SNS's occurrence and duration were usually not associated with the same timescale. Discussion: Our results suggest that different time-varying factors modulate within-patient seizure evolutions over multiple timescales, with separate processes modulating a SNS's occurrence and duration. These findings imply that the development of time-adaptive treatments in epilepsy must account for several separate properties of epileptic seizures, and similar principles likely apply to other neurological conditions.

q-bio.NC↗

arXiv2023

Data Encoding For Healthcare Data Democratisation and Information Leakage Prevention

The lack of data democratization and information leakage from trained models hinder the development and acceptance of robust deep learning-based healthcare solutions. This paper argues that irreversible data encoding can provide an effective solution to achieve data democratization without violating the privacy constraints imposed on healthcare data and clinical models. An ideal encoding framework transforms the data into a new space where it is imperceptible to a manual or computational inspection. However, encoded data should preserve the semantics of the original data such that deep learning models can be trained effectively. This paper hypothesizes the characteristics of the desired encoding framework and then exploits random projections and random quantum encoding to realize this framework for dense and longitudinal or time-series data. Experimental evaluation highlights that models trained on encoded time-series data effectively uphold the information bottleneck principle and hence, exhibit lesser information leakage from trained models.

cs.LG↗

arXiv2023

Identifying epileptogenic abnormalities through spatial clustering of MEG interictal band power

Successful epilepsy surgery depends on localising and resecting cerebral abnormalities and networks that generate seizures. Abnormalities, however, may be widely distributed across multiple discontiguous areas. We propose spatially constrained clusters as candidate areas for further investigation, and potential resection. We quantified the spatial overlap between the abnormality cluster and subsequent resection, hypothesising a greater overlap in seizure-free patients. Thirty-four individuals with refractory focal epilepsy underwent pre-surgical resting-state interictal MEG recording. Fourteen individuals were totally seizure free (ILAE 1) after surgery and 20 continued to have some seizures post-operatively (ILAE 2+). Band power abnormality maps were derived using controls as a baseline. Patient abnormalities were spatially clustered using the k-means algorithm. The tissue within the cluster containing the most abnormal region was compared with the resection volume using the dice score. The proposed abnormality cluster overlapped with the resection in 71% of ILAE 1 patients. Conversely, an overlap only occurred in 15% of ILAE 2+ patients. This effect discriminated outcome groups well (AUC=0.82). Our novel approach identifies clusters of spatially similar tissue with high abnormality. This is clinically valuable, providing (i) a data-driven framework to validate current hypotheses of the epileptogenic zone localisation or (ii) to guide further investigation.

q-bio.NC↗

arXiv2023

Interictal MEG abnormalities to guide intracranial electrode implantation and predict surgical outcome

Intracranial EEG (iEEG) is the gold standard technique for epileptogenic zone (EZ) localisation, but requires a hypothesis of which tissue is epileptogenic, guided by qualitative analysis of seizure semiology and other imaging modalities such as magnetoencephalography (MEG). We hypothesised that if quantifiable MEG band power abnormalities were sampled by iEEG, then patients' post-resection seizure outcome were better. Thirty-two individuals with neocortical epilepsy underwent MEG and iEEG recordings as part of pre-surgical evaluation. Interictal MEG band power abnormalities were derived using 70 healthy controls as a normative baseline. MEG abnormality maps were compared to electrode implantation, with the spatial overlap of iEEG electrodes and MEG abnormalities recorded. Finally, we assessed if the implantation of electrodes in abnormal tissue, and resection of the strongest abnormalities determined by MEG and iEEG explained surgical outcome. Intracranial electrodes were implanted in brain tissue with the most abnormal MEG findings in individuals that were seizure-free post-resection (T=3.9, p=0.003). The overlap between MEG abnormalities and iEEG electrodes distinguished outcome groups moderately well (AUC=0.68). In isolation, the resection of the strongest MEG and iEEG abnormalities separated surgical outcome groups well (AUC=0.71, AUC=0.74 respectively). A model incorporating all three features separated outcome groups best (AUC=0.80). Intracranial EEG is a key tool to delineate the EZ and help render patients seizure-free after resection. We showed that data-driven abnormalities derived from interictal MEG recordings have clinical value and may help guide electrode placement in individuals with neocortical epilepsy. Finally, our predictive model of post-operative seizure-freedom, which leverages both MEG and iEEG recordings, may aid patient counselling of expected outcome.

q-bio.NC↗

arXiv2023

Normative brain mapping using scalp EEG and potential clinical application

A normative electrographic activity map could be a powerful resource to understand normal brain function and identify abnormal activity. Here, we present a normative brain map using scalp EEG in terms of relative band power. In this exploratory study we investigate its temporal stability, its similarity to other imaging modalities, and explore a potential clinical application. We constructed scalp EEG normative maps of brain dynamics from 17 healthy controls using source-localised resting-state scalp recordings. We then correlated these maps with those acquired from MEG and intracranial EEG to investigate their similarity. Lastly, we use the normative maps to lateralise abnormal regions in epilepsy. Spatial patterns of band powers were broadly consistent with previous literature and stable across recordings. Scalp EEG normative maps were most similar to other modalities in the alpha band, and relatively similar across most bands. Towards a clinical application in epilepsy, we found abnormal temporal regions ipsilateral to the epileptogenic hemisphere. Scalp EEG relative band power normative maps are spatially stable across time, in keeping with MEG and intracranial EEG results. Normative mapping is feasible and may be potentially clinically useful in epilepsy. Future studies with larger sample sizes and high-density EEG are now required for validation.

q-bio.NC↗

arXiv2023

Self-supervised Video-centralised Transformer for Video Face Clustering

This paper presents a novel method for face clustering in videos using a video-centralised transformer. Previous works often employed contrastive learning to learn frame-level representation and used average pooling to aggregate the features along the temporal dimension. This approach may not fully capture the complicated video dynamics. In addition, despite the recent progress in video-based contrastive learning, few have attempted to learn a self-supervised clustering-friendly face representation that benefits the video face clustering task. To overcome these limitations, our method employs a transformer to directly learn video-level representations that can better reflect the temporally-varying property of faces in videos, while we also propose a video-centralised self-supervised framework to train the transformer model. We also investigate face clustering in egocentric videos, a fast-emerging field that has not been studied yet in works related to face clustering. To this end, we present and release the first large-scale egocentric video face clustering dataset named EasyCom-Clustering. We evaluate our proposed method on both the widely used Big Bang Theory (BBT) dataset and the new EasyCom-Clustering dataset. Results show the performance of our video-centralised transformer has surpassed all previous state-of-the-art methods on both benchmarks, exhibiting a self-attentive understanding of face videos.

cs.CV↗

arXiv2023

MEG abnormalities and mechanisms of surgical failure in neocortical epilepsy

Neocortical epilepsy surgery fails to achieve post-operative seizure freedom in 30-40% of cases. It is not fully understood why surgery in some patients is unsuccessful. Comparing interictal MEG bandpower from patients to normative maps, which describe healthy spatial and population variability, we identify patient specific abnormalities relating to surgical failure. We propose three mechanisms contributing to poor surgical outcome; 1) failure to resect abnormalities, 2) failing to remove all epileptogenic abnormalities, and 3) insufficiently impacting the overall cortical abnormality. We develop markers of these mechanisms, validating them against patient outcomes. Resting-state MEG data were acquired for 70 healthy controls and 32 patients with refractory neocortical epilepsy. Relative bandpower maps were computed using source localised recordings from healthy controls. Patient and region-specific bandpower abnormalities were estimated as the maximum absolute z-score, using healthy data as a baseline. Resected regions were identified from post-operative MRI. We hypothesised our mechanism markers would discriminate patient's post-surgery seizure outcomes. Mechanisms of surgical failure discriminate surgical outcome groups (Abnormalities not targeted: AUC=0.80, Partial resection of the epileptogenic zone: AUC=0.68, Insufficient cortical abnormality impact: AUC=0.64). Leveraging all markers together found that 95% of those who were not seizure free had markers of surgical failure in at least one of the three proposed mechanisms. In contrast, of those patients markers for any mechanism, 80% were seizure-free. Abnormality mapping across the brain is important for a wide range of neurological conditions. Here we demonstrated that interictal MEG bandpower mapping has merit for localising pathology and improving our mechanistic understanding of epilepsy.

q-bio.NC↗

arXiv2023

Temporal stability of intracranial EEG abnormality maps for localising epileptogenic tissue

Objective: Identifying abnormalities in interictal intracranial EEG, by comparing patient data to a normative map, has shown promise for the localisation of epileptogenic tissue and prediction of outcome. The approach typically uses short interictal segments of around one minute. However, the temporal stability of findings has not been established. Methods: Here, we generated a normative map of iEEG in non-pathological brain tissue from 249 patients. We computed regional band power abnormalities in a separate cohort of 39 patients for the duration of their monitoring period (0.92-8.62 days of iEEG data, mean 4.58 days per patient, over 4,800 hours recording). To assess the localising value of band power abnormality, we computed DRS - a measure of how different the surgically resected and spared tissue were in terms of band power abnormalities - over time. Results: In each patient, band power abnormality was relatively consistent over time. The median DRS of the entire recording period separated seizure free (ILAE = 1) and not seizure free (ILAE > 1) patients well (AUC = 0.69). This effect was similar interictally (AUC = 0.69) and peri-ictally (AUC = 0.71). Significance: Our results suggest that band power abnormality DRS, as a predictor of outcomes from epilepsy surgery, is a relatively robust metric over time. These findings add further support for abnormality mapping of neurophysiology data during presurgical evaluation.

q-bio.NC↗

arXiv2023

A library of quantitative markers of seizure severity

Purpose: Understanding fluctuations of seizure severity within individuals is important for defining treatment outcomes and response to therapy, as well as developing novel treatments for epilepsy. Current methods for grading seizure severity rely on qualitative interpretations from patients and clinicians. Quantitative measures of seizure severity would complement existing approaches, for EEG monitoring, outcome monitoring, and seizure prediction. Therefore, we developed a library of quantitative electroencephalographic (EEG) markers that assess the spread and intensity of abnormal electrical activity during and after seizures. Methods: We analysed intracranial EEG (iEEG) recordings of 1056 seizures from 63 patients. For each seizure, we computed 16 markers of seizure severity that capture the signal magnitude, spread, duration, and post-ictal suppression of seizures. Results: Quantitative EEG markers of seizure severity distinguished focal vs. subclinical and focal vs. FTBTC seizures across patients. In individual patients, 71% had a moderate to large difference (ranksum r > 0.3) between focal and subclinical seizures in three or more markers. Circadian and longer-term changes in severity were found for 67% and 53% of patients, respectively. Conclusion: We demonstrate the feasibility of using quantitative iEEG markers to measure seizure severity. Our quantitative markers distinguish between seizure types and are therefore sensitive to established qualitative differences in seizure severity. Our results also suggest that seizure severity is modulated over different timescales. We envisage that our proposed seizure severity library will be expanded and updated in collaboration with the epilepsy research community to include more measures and modalities.

q-bio.NC↗