Dataset Information

Unsupervised machine-learning classification of electrophysiologically active electrodes during human cognitive task performance.

ABSTRACT: Identification of active electrodes that record task-relevant neurophysiological activity is needed for clinical and industrial applications as well as for investigating brain functions. We developed an unsupervised, fully automated approach to classify active electrodes showing event-related intracranial EEG (iEEG) responses from 115 patients performing a free recall verbal memory task. Our approach employed new interpretable metrics that quantify spectral characteristics of the normalized iEEG signal based on power-in-band and synchrony measures. Unsupervised clustering of the metrics identified distinct sets of active electrodes across different subjects. In the total population of 11,869 electrodes, our method achieved 97% sensitivity and 92.9% specificity with the most efficient metric. We validated our results with anatomical localization revealing significantly greater distribution of active electrodes in brain regions that support verbal memory processing. We propose our machine-learning framework for objective and efficient classification and interpretation of electrophysiological signals of brain activities supporting memory and cognition.

SUBMITTER: Saboo KV

PROVIDER: S-EPMC6874617 | biostudies-literature | 2019 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Unsupervised machine-learning classification of electrophysiologically active electrodes during human cognitive task performance.

Saboo Krishnakant V KV Varatharajah Yogatheesan Y Berry Brent M BM Kremen Vaclav V Sperling Michael R MR Davis Kathryn A KA Jobst Barbara C BC Gross Robert E RE Lega Bradley B Sheth Sameer A SA Worrell Gregory A GA Iyer Ravishankar K RK Kucewicz Michal T MT

Scientific reports 20191122 1

Identification of active electrodes that record task-relevant neurophysiological activity is needed for clinical and industrial applications as well as for investigating brain functions. We developed an unsupervised, fully automated approach to classify active electrodes showing event-related intracranial EEG (iEEG) responses from 115 patients performing a free recall verbal memory task. Our approach employed new interpretable metrics that quantify spectral characteristics of the normalized iEEG ...[more]

PMID: 31758077

Similar Datasets

Project description:Machine learning (ML) methods have great potential to transform chemical discovery by accelerating the exploration of chemical space and drawing scientific insights from data. However, modern chemical reaction ML models, such as those based on graph neural networks (GNNs), must be trained on a large amount of labelled data in order to avoid overfitting the data and thus possessing low accuracy and transferability. In this work, we propose a strategy to leverage unlabelled data to learn accurate ML models for small labelled chemical reaction data. We focus on an old and prominent problem-classifying reactions into distinct families-and build a GNN model for this task. We first pretrain the model on unlabelled reaction data using unsupervised contrastive learning and then fine-tune it on a small number of labelled reactions. The contrastive pretraining learns by making the representations of two augmented versions of a reaction similar to each other but distinct from other reactions. We propose chemically consistent reaction augmentation methods that protect the reaction center and find they are the key for the model to extract relevant information from unlabelled data to aid the reaction classification task. The transfer learned model outperforms a supervised model trained from scratch by a large margin. Further, it consistently performs better than models based on traditional rule-driven reaction fingerprints, which have long been the default choice for small datasets, as well as those based on reaction fingerprints derived from masked language modelling. In addition to reaction classification, the effectiveness of the strategy is tested on regression datasets; the learned GNN-based reaction fingerprints can also be used to navigate the chemical reaction space, which we demonstrate by querying for similar reactions. The strategy can be readily applied to other predictive reaction problems to uncover the power of unlabelled data for learning better models with a limited supply of labels.

Project description:Background: Frequent simulation-based education is recommended to improve health outcomes during neonatal resuscitation but is often inaccessible due to time, resource, and personnel requirements. Digital simulation presents a potential alternative; however, its effectiveness and reception by healthcare professionals (HCPs) remains largely unexplored. Objectives: This study explores HCPs' attitudes toward a digital simulator, technology, and mindset to elucidate their effects on neonatal resuscitation performance in simulation-based assessments. Methods: The study was conducted from April to August 2019 with 2-month (June-October 2019) and 5-month (September 2019-January 2020) follow-up at a tertiary perinatal center in Edmonton, Canada. Of 300 available neonatal HCPs, 50 participated. Participants completed a demographic survey, a pretest, two practice scenarios using the RETAIN neonatal resuscitation digital simulation, a posttest, and an attitudinal survey (100% response rate). Participants repeated the posttest scenario in 2 months (86% response rate) and completed another posttest scenario using a low-fidelity, tabletop simulator (80% response rate) 5 months after the initial study intervention. Participants' survey responses were collected to measure attitudes toward digital simulation and technology. Knowledge was assessed at baseline (pretest), acquisition (posttest), retention (2-month posttest), and transfer (5-month posttest). Results: Fifty neonatal HCPs participated in this study (44 females and 6 males; 27 nurses, 3 nurse practitioners, 14 respiratory therapists, and 6 doctors). Most participants reported technology in medical education as useful and beneficial. Three attitudinal clusters were identified by a hierarchical clustering algorithm based on survey responses. Although participants exhibited diverse attitudinal paths, they all improved neonatal resuscitation performance after using the digital simulator and successfully transferred their knowledge to a new medium. Conclusions: Digital simulation improved HCPs' neonatal resuscitation performance. Medical education may benefit by incorporating technology during simulation training.

Dataset Information

Unsupervised machine-learning classification of electrophysiologically active electrodes during human cognitive task performance.

Publications

Unsupervised machine-learning classification of electrophysiologically active electrodes during human cognitive task performance.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets