Dataset Information

EEG-based outcome prediction after cardiac arrest with convolutional neural networks: Performance and visualization of discriminative features.

ABSTRACT: Prognostication for comatose patients after cardiac arrest is a difficult but essential task. Currently, visual interpretation of electroencephalogram (EEG) is one of the main modality used in outcome prediction. There is a growing interest in computer-assisted EEG interpretation, either to overcome the possible subjectivity of visual interpretation, or to identify complex features of the EEG signal. We used a one-dimensional convolutional neural network (CNN) to predict functional outcome based on 19-channel-EEG recorded from 267 adult comatose patients during targeted temperature management after CA. The area under the receiver operating characteristic curve (AUC) on the test set was 0.885. Interestingly, model architecture and fine-tuning only played a marginal role in classification performance. We then used gradient-weighted class activation mapping (Grad-CAM) as visualization technique to identify which EEG features were used by the network to classify an EEG epoch as favorable or unfavorable outcome, and also to understand failures of the network. Grad-CAM showed that the network relied on similar features than classical visual analysis for predicting unfavorable outcome (suppressed background, epileptiform transients). This study confirms that CNNs are promising models for EEG-based prognostication in comatose patients, and that Grad-CAM can provide explanation for the models' decision-making, which is of utmost importance for future use of deep learning models in a clinical setting.

SUBMITTER: Jonas S

PROVIDER: S-EPMC6865376 | biostudies-literature | 2019 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

EEG-based outcome prediction after cardiac arrest with convolutional neural networks: Performance and visualization of discriminative features.

Jonas Stefan S Rossetti Andrea O AO Oddo Mauro M Jenni Simon S Favaro Paolo P Zubler Frederic F

Human brain mapping 20190719 16

Prognostication for comatose patients after cardiac arrest is a difficult but essential task. Currently, visual interpretation of electroencephalogram (EEG) is one of the main modality used in outcome prediction. There is a growing interest in computer-assisted EEG interpretation, either to overcome the possible subjectivity of visual interpretation, or to identify complex features of the EEG signal. We used a one-dimensional convolutional neural network (CNN) to predict functional outcome based ...[more]

PMID: 31322793

Similar Datasets

Project description:PurposeWe aimed to determine which early EEG features and feature combinations most accurately predicted short-term neurobehavioral outcomes and survival in children resuscitated after cardiac arrest.MethodsThis was a prospective, single-center observational study of infants and children resuscitated from cardiac arrest who underwent conventional EEG monitoring with standardized EEG scoring. Logistic regression evaluated the marginal effect of each EEG variable or EEG variable combinations on the outcome. The primary outcome was neurobehavioral outcome (Pediatric Cerebral Performance Category score), and the secondary outcome was mortality. The authors identified the models with the highest areas under the receiver operating characteristic curve (AUC), evaluated the optimal models using a 5-fold cross-validation approach, and calculated test characteristics maximizing specificity.ResultsEighty-nine infants and children were evaluated. Unfavorable neurologic outcome (Pediatric Cerebral Performance Category score 4-6) occurred in 44 subjects (49%), including mortality in 30 subjects (34%). A model incorporating a four-level EEG Background Category (normal, slow-disorganized, discontinuous or burst-suppression, or attenuated-flat), stage 2 Sleep Transients (present or absent), and Reactivity-Variability (present or absent) had the highest AUC. Five-fold cross-validation for the optimal model predicting neurologic outcome indicated a mean AUC of 0.75 (range, 0.70-0.81) and for the optimal model predicting mortality indicated a mean AUC of 0.84 (range, 0.76-0.97). The specificity for unfavorable neurologic outcome and mortality were 95% and 97%, respectively. The positive predictive value for unfavorable neurologic outcome and mortality were both 86%.ConclusionsThe specificity of the optimal model using a combination of early EEG features was high for unfavorable neurologic outcome and mortality in critically ill children after cardiac arrest. However, the positive predictive value was only 86% for both outcomes. Therefore, EEG data must be considered in overall clinical context when used for neuroprognostication early after cardiac arrest.

Project description:Motivation:In addition to substitution frequency data from protein sequence alignments, many state-of-the-art methods for contact prediction rely on additional sources of information, or features, of protein sequences in order to predict residue-residue contacts, such as solvent accessibility, predicted secondary structure, and scores from other contact prediction methods. It is unclear how much of this information is needed to achieve state-of-the-art results. Here, we show that using deep neural network models, simple alignment statistics contain sufficient information to achieve state-of-the-art precision. Our prediction method, DeepCov, uses fully convolutional neural networks operating on amino-acid pair frequency or covariance data derived directly from sequence alignments, without using global statistical methods such as sparse inverse covariance or pseudolikelihood estimation. Results:Comparisons against CCMpred and MetaPSICOV2 show that using pairwise covariance data calculated from raw alignments as input allows us to match or exceed the performance of both of these methods. Almost all of the achieved precision is obtained when considering relatively local windows (around 15 residues) around any member of a given residue pairing; larger window sizes have comparable performance. Assessment on a set of shallow sequence alignments (fewer than 160 effective sequences) indicates that the new method is substantially more precise than CCMpred and MetaPSICOV2 in this regime, suggesting that improved precision is attainable on smaller sequence families. Overall, the performance of DeepCov is competitive with the state of the art, and our results demonstrate that global models, which employ features from all parts of the input alignment when predicting individual contacts, are not strictly needed in order to attain precise contact predictions. Availability and implementation:DeepCov is freely available at https://github.com/psipred/DeepCov. Supplementary information:Supplementary data are available at Bioinformatics online.

Project description:One of the main obstacles for the implementation of deep convolutional neural networks (DCNNs) in the clinical pathology workflow is their low capability to overcome variability in slide preparation and scanner configuration, that leads to changes in tissue appearance. Some of these variations may not be not included in the training data, which means that the models have a risk to not generalize well. Addressing such variations and evaluating them in reproducible scenarios allows understanding of when the models generalize better, which is crucial for performance improvements and better DCNN models. Staining normalization techniques (often based on color deconvolution and deep learning) and color augmentation approaches have shown improvements in the generalization of the classification tasks for several tissue types. Domain-invariant training of DCNN's is also a promising technique to address the problem of training a single model for different domains, since it includes the source domain information to guide the training toward domain-invariant features, achieving state-of-the-art results in classification tasks. In this article, deep domain adaptation in convolutional networks (DANN) is applied to computational pathology and compared with widely used staining normalization and color augmentation methods in two challenging classification tasks. The classification tasks rely on two openly accessible datasets, targeting Gleason grading in prostate cancer, and mitosis classification in breast tissue. The benchmark of the different techniques and their combination in two DCNN architectures allows us to assess the generalization abilities and advantages of each method in the considered classification tasks. The code for reproducing our experiments and preprocessing the data is publicly available. Quantitative and qualitative results show that the use of DANN helps model generalization to external datasets. The combination of several techniques to manage color heterogeneity suggests that several methods together, such as color augmentation methods with DANN training, can generalize even further. The results do not show a single best technique among the considered methods, even when combining them. However, color augmentation and DANN training obtain most often the best results (alone or combined with color normalization and color augmentation). The statistical significance of the results and the embeddings visualizations provide useful insights to design DCNN that generalizes to unseen staining appearances. Furthermore, in this work, we release for the first time code for DANN evaluation in open access datasets for computational pathology. This work opens the possibility for further research on using DANN models together with techniques that can overcome the tissue preparation differences across datasets to tackle limited generalization.

Dataset Information

EEG-based outcome prediction after cardiac arrest with convolutional neural networks: Performance and visualization of discriminative features.

Publications

EEG-based outcome prediction after cardiac arrest with convolutional neural networks: Performance and visualization of discriminative features.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets