Convolutional neural networks ensemble model for neonatal seizure detection

Deep learning-based auditory attention decoding in listeners with hearing impairment. J Neural Eng 2024;21:036022. [PMID: 38729132 DOI: 10.1088/1741-2552/ad49d7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Objective.This study develops a deep learning (DL) method for fast auditory attention decoding (AAD) using electroencephalography (EEG) from listeners with hearing impairment (HI). It addresses three classification tasks: differentiating noise from speech-in-noise, classifying the direction of attended speech (left vs. right) and identifying the activation status of hearing aid noise reduction algorithms (OFF vs. ON). These tasks contribute to our understanding of how hearing technology influences auditory processing in the hearing-impaired population.Approach.Deep convolutional neural network (DCNN) models were designed for each task. Two training strategies were employed to clarify the impact of data splitting on AAD tasks: inter-trial, where the testing set used classification windows from trials that the training set had not seen, and intra-trial, where the testing set used unseen classification windows from trials where other segments were seen during training. The models were evaluated on EEG data from 31 participants with HI, listening to competing talkers amidst background noise.Main results.Using 1 s classification windows, DCNN models achieve accuracy (ACC) of 69.8%, 73.3% and 82.9% and area-under-curve (AUC) of 77.2%, 80.6% and 92.1% for the three tasks respectively on inter-trial strategy. In the intra-trial strategy, they achieved ACC of 87.9%, 80.1% and 97.5%, along with AUC of 94.6%, 89.1%, and 99.8%. Our DCNN models show good performance on short 1 s EEG samples, making them suitable for real-world applications. Conclusion: Our DCNN models successfully addressed three tasks with short 1 s EEG windows from participants with HI, showcasing their potential. While the inter-trial strategy demonstrated promise for assessing AAD, the intra-trial approach yielded inflated results, underscoring the important role of proper data splitting in EEG-based AAD tasks.Significance.Our findings showcase the promising potential of EEG-based tools for assessing auditory attention in clinical contexts and advancing hearing technology, while also promoting further exploration of alternative DL architectures and their potential constraints.

Unified Convolutional Sparse Transformer for Disease Diagnosis, Monitoring, Drug Development, and Therapeutic Effect Prediction from EEG Raw Data

Electroencephalogram (EEG) analysis plays an indispensable role across contemporary medical applications, which encompasses diagnosis, monitoring, drug discovery, and therapeutic assessment. This work puts forth an end-to-end deep learning framework that is uniquely tailored for versatile EEG analysis tasks by directly operating on raw waveform inputs. It aims to address the challenges of manual feature engineering and the neglect of spatial interrelationships in existing methodologies. Specifically, a spatial channel attention module is introduced to emphasize the critical inter-channel dependencies in EEG signals through channel statistics aggregation and multi-layer perceptron operations. Furthermore, a sparse transformer encoder is used to leverage selective sparse attention in order to efficiently process long EEG sequences while reducing computational complexity. Distilling convolutional layers further concatenates the temporal features and retains only the salient patterns. As it was rigorously evaluated on key EEG datasets, our model consistently accomplished a superior performance over the current approaches in detection and classification assignments. By accounting for both spatial and temporal relationships in an end-to-end paradigm, this work facilitates a versatile, automated EEG understanding across diseases, subjects, and objectives through a singular yet customizable architecture. Extensive empirical validation and further architectural refinement may promote broader clinical adoption prospects.

A robust seizure detection and prediction method with feature selection and spatio-temporal casual neural network model

Objective.Epilepsy is a fairly common condition that affects the brain and causes frequent seizures. The sudden and recurring epilepsy brings a series of safety hazards to patients, which seriously affects the quality of their life. Therefore, real-time diagnosis of electroencephalogram (EEG) in epilepsy patients is of great significance. However, the conventional methods take in a tremendous amount of features to train the models, resulting in high computation cost and low portability. Our objective is to propose an efficient, light and robust seizure detecting and predicting algorithm.Approach.The algorithm is based on an interpretative feature selection method and spatial-temporal causal neural network (STCNN). The feature selection method eliminates the interference factors between different features and reduces the model size and training difficulties. The STCNN model takes both temporal and spatial information to accurately and dynamically track and diagnose the changing of the features. Considering the differences between medical application scenarios and patients, leave-one-out cross validation (LOOCV) and cross-patient validation (CPV) methods are used to conduct experiments on the dataset collected at the Children's Hospital Boston (CHB-MIT), Siena and Kaggle competition datasets.Main results.In LOOCV-based method, the detection accuracy and prediction sensitivity have been improved. A significant improvement is also achieved in the CPV-based method.Significance.The experimental results show that our proposed algorithm exhibits superior performance and robustness in seizure detection and prediction, which indicates it has higher capability to deal with different and complicated clinical situations.

A machine learning approach to seizure detection in a rat model of post-traumatic epilepsy

Epilepsy is a common neurologic condition frequently investigated using rodent models, with seizures identified by electroencephalography (EEG). Given technological advances, large datasets of EEG are widespread and amenable to machine learning approaches for identification of seizures. While such approaches have been explored for human EEGs, machine learning approaches to identifying seizures in rodent EEG are limited. We utilized a predesigned deep convolutional neural network (DCNN), GoogLeNet, to classify images for seizure identification. Training images were generated through multiplexing spectral content (scalograms), kurtosis, and entropy for two-second EEG segments. Over 2200 h of EEG data were scored for the presence of seizures, with 95.6% of seizures identified by the DCNN and a false positive rate of 34.2% (1.52/h), as compared to visual scoring. Multiplexed images were superior to scalograms alone (scalogram-kurtosis-entropy 0.956 ± 0.010, scalogram 0.890 ± 0.028, t(7) = 3.54, p < 0.01) and a DCNN trained specifically for the individual animal was superior to using DCNNs across animals (intra-animal 0.960 ± 0.0094, inter-animal 0.811 ± 0.015, t(30) = 5.54, p < 0.01). For this dataset the DCNN approach is superior to a previously described algorithm utilizing longer local line lengths, calculated from wavelet-decomposition of EEG, to identify seizures. We demonstrate the novel use of a predesigned DCNN constructed to classify images, utilizing multiplexed images of EEG spectral content, kurtosis, and entropy, to rapidly and objectively identifies seizures in a large dataset of rat EEG with high sensitivity.

A Modified Aquila-Based Optimized XGBoost Framework for Detecting Probable Seizure Status in Neonates

Electroencephalography (EEG) is increasingly being used in pediatric neurology and provides opportunities to diagnose various brain illnesses more accurately and precisely. It is thought to be one of the most effective tools for identifying newborn seizures, especially in Neonatal Intensive Care Units (NICUs). However, EEG interpretation is time-consuming and requires specialists with extensive training. It can be challenging and time-consuming to distinguish between seizures since they might have a wide range of clinical characteristics and etiologies. Technological advancements such as the Machine Learning (ML) approach for the rapid and automated diagnosis of newborn seizures have increased in recent years. This work proposes a novel optimized ML framework to eradicate the constraints of conventional seizure detection techniques. Moreover, we modified a novel meta-heuristic optimization algorithm (MHOA), named Aquila Optimization (AO), to develop an optimized model to make our proposed framework more efficient and robust. To conduct a comparison-based study, we also examined the performance of our optimized model with that of other classifiers, including the Decision Tree (DT), Random Forest (RF), and Gradient Boosting Classifier (GBC). This framework was validated on a public dataset of Helsinki University Hospital, where EEG signals were collected from 79 neonates. Our proposed model acquired encouraging results showing a 93.38% Accuracy Score, 93.9% Area Under the Curve (AUC), 92.72% F1 score, 65.17% Kappa, 93.38% sensitivity, and 77.52% specificity. Thus, it outperforms most of the present shallow ML architectures by showing improvements in accuracy and AUC scores. We believe that these results indicate a major advance in the detection of newborn seizures, which will benefit the medical community by increasing the reliability of the detection process.

Design of a Multi-Feature Classification Scheme for Infant Epileptic Seizures

Neonatal epileptic seizures take place in the early childhood years, accounting for a severe condition with several deaths and neurological problems in newborn neonates. Despite the early advancements on the diagnosis and/or treatment of this condition, as a major difficulty accounts the inability of the physicians to identify and characterize a seizure, as one a small percentage gets detected in neonatal intensive care units (NICU). An important step towards any kind of seizure classification is the detection and reduction of non-cerebral activity. Towards this direction, our multi-feature approach contains spectral and statistical characteristics of EEG signals of 79 infants with suspicion of seizure and assesses the performance of two classification algorithms iteratively. The trained models (Support Vector Machine (SVM) and Random Forest classifiers) yielded high classification performance (>80% and >85% respectively). A robust neonatal seizure classification scheme is thus proposed, along with nine high scoring spectrum and statistical features. The importance of embedding an artefact reduction approach is also discussed, since the complex artifacts spread throughout the signals have great impact on the accuracy of the algorithms. The nine extracted high scoring spectral and statistical features might be used as potential biomarkers for neonatal seizure prediction in a clinical setting.

Neonatal Seizure Detection Using a Wearable Multi-Sensor System

Neonatal seizure is an important clinical symptom of brain dysfunction, which is more common in infancy than in childhood. At present, video electroencephalogram (VEEG) technology is widely used in clinical practice. However, video electroencephalogram technology has several disadvantages. For example, the wires connecting the medical instruments may interfere with the infant's movement and the gel patch electrode or disk electrode commonly used to monitor EEG may cause skin allergies or even tears. For the above reasons, we developed a wearable multi-sensor platform for newborns to collect physiological and movement signals. In this study, we designed a second-generation multi-sensor platform and developed an automatic detection algorithm for neonatal seizures based on ECG, respiration and acceleration. Data for 38 neonates were recorded at the Children's Hospital of Fudan University in Shanghai. The total recording time was approximately 300 h. Four of the patients had seizures during data collection. The total recording time for the four patients was approximately 34 h, with 30 seizure episodes recorded. These data were evaluated by the algorithm. To evaluate the effectiveness of combining ECG, respiration and movement, we compared the performance of three types of seizure detectors. The first detector included features from ECG, respiration and acceleration records; the second detector incorporated features based on respiratory movement from respiration and acceleration records; and the third detector used only ECG-based features from ECG records. Our study illustrated that, compared with the detector utilizing individual modal features, multi-modal feature detectors could achieve favorable overall performance, reduce false alarm rates and give higher F-measures.

A graph convolutional neural network for the automated detection of seizures in the neonatal EEG

BACKGROUND AND OBJECTIVE

Neonatal seizures are the most common clinical presentation of neurological conditions and can have adverse effects on the neurodevelopment of the neonatal brain. Visual detection of these events from continuous EEG recordings is a laborious and time-consuming task. We propose a novel algorithm for the automated detection of neonatal seizures.

METHODS

In this study, we propose a novel deep learning model based on Graph Convolutional Neural Networks for the automated detection of neonatal seizures. Unlike other methods exploiting mainly the temporal information contained in EEG signals, our method also considers long-range spatial information, i.e., the interdependencies across EEG signals. The temporal information is embedded as graph signals in the graph representation of the EEG recordings and includes EEG features extracted from the EEG signals in the time and frequency domains. The spatial information is represented as functional connections among the EEG channels (calculated by the phase-locking value and the mean squared coherence) or as maps of Euclidean distances. These different spatial representations were evaluated to assess their efficiency in providing more discriminative features for an effective detection of neonatal seizures. The model performance was assessed on a publicly available dataset of continuous EEG signals recorded from 39 neonates by means of the area under the curve (AUC) and the AUC for specificity values greater than 90% (AUC90).

RESULTS

After applying post-processing, consisting in smoothing the output of the classifiers, the models based on the mean squared coherence, the phase-locking value, and the Euclidean distance respectively reached a median AUC of 99.1% (IQR: 96.8%-99.6%), 99% (IQR: 95.2%-99.7%), and 97.3% (IQR: 86.3%-99.6%), and a median AUC90 of 96%, 95.7%, and 94.9%. These values are superior or comparable to those reached by methods considered as state-of-the-art in this field.

CONCLUSIONS

Our results show that the EEG graph representations drawn from functional connectivity measures can effectively leverage interdependencies among EEG signals and lead to reliable detection of neonatal seizures. Furthermore, our model has the advantage of requiring only temporal annotations on seizures for the training phase, making it more appealing for clinical applications.

Epileptic Seizure Detection Based on Bidirectional Gated Recurrent Unit Network

Visual inspection of long-term electroencephalography (EEG) is a tedious task for physicians in neurology. Based on bidirectional gated recurrent unit (Bi-GRU) neural network, an automatic seizure detection method is proposed in this paper to facilitate the diagnosis and treatment of epilepsy. Firstly, wavelet transforms are applied to EEG recordings for filtering pre-processing. Then the relative energies of signals in several particular frequency bands are calculated and inputted into Bi-GRU network. Afterwards, the outputs of Bi-GRU network are further processed by moving average filtering, threshold comparison and seizure merging to generate the discriminant results that the tested EEG belong to seizure or not. Evaluated on CHB-MIT scalp EEG database, the proposed seizure detection method obtained an average sensitivity of 93.89% and an average specificity of 98.49%. 124 out of 128 seizures were correctly detected and the achieved average false detection rate was 0.31 per hour on 867.14 h testing data. The results show the superiority of Bi-GRU network in seizure detection and the proposed detection method has a promising potential in the monitoring of long-term EEG.

Ensemble Learning Using Individual Neonatal Data for Seizure Detection

Objective: Sharing medical data between institutions is difficult in practice due to data protection laws and official procedures within institutions. Therefore, most existing algorithms are trained on relatively small electroencephalogram (EEG) data sets which is likely to be detrimental to prediction accuracy. In this work, we simulate a case when the data can not be shared by splitting the publicly available data set into disjoint sets representing data in individual institutions. Methods and procedures: We propose to train a (local) detector in each institution and aggregate their individual predictions into one final prediction. Four aggregation schemes are compared, namely, the majority vote, the mean, the weighted mean and the Dawid-Skene method. The method was validated on an independent data set using only a subset of EEG channels. Results: The ensemble reaches accuracy comparable to a single detector trained on all the data when sufficient amount of data is available in each institution. Conclusion: The weighted mean aggregation scheme showed best performance, it was only marginally outperformed by the Dawid–Skene method when local detectors approach performance of a single detector trained on all available data. Clinical impact: Ensemble learning allows training of reliable algorithms for neonatal EEG analysis without a need to share the potentially sensitive EEG data between institutions.