Seizure Detection Using Time Delay Neural Networks and LSTMs

Thyagachandran, Anand; Kumar, Mari Ganesh; Sur, Mriganka; Aghoram, Rajeswari; Murthy, Hema A.

doi:10.1109/spmb50085.2020.9353636

Cited by 7 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, when the 2D models were trained and evaluated on the CHB-MIT dataset, we obtained the worst results thus far, with PRE lower than 5% for all cases. These numerical results are in line with many 2D models reported in the literature 20,28,29,[57][58][59] .…”

Section: Benefits Of Channel-level Seizure Detectionsupporting

confidence: 91%

“…In recent studies on automated seizure detection from EEG, the seizure detectors are validated mainly on two public seizure datasets: the Temple University Hospital seizure (TUH-SZ) dataset [18][19][20][21] and the Children's Hospital Boston Massachusetts Institute of Technology (CHB-MIT) dataset 19,[22][23][24] . In many studies, different seizure detectors are proposed, including (standard) machine learning models 25,26 , convolutional neural networks (CNNs) 18,21,22,24,27 , recurrent neural networks (RNNs) 18,20 , long short-term memory (LSTM) 28,29 , transformer 30,31 , transfer learning models 20,[32][33][34] , and temporal graph convolutional networks (TGCNs) 35 . The seizure detectors proposed in these studies are similar in architecture or implementation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi-center Assessment of CNN-Transformer with Belief Matching Loss for Patient-independent Seizure Detection in Scalp and Intracranial EEG

Peh

Thangavel

Yao

et al. 2022

Preprint

View full text Add to dashboard Cite

Neurologists typically identify epileptic seizures from electroencephalograms (EEGs) by visual inspection. This process is often time-consuming, especially for EEG recordings that last several hours or even days. To expedite the process, a reliable, automated, and patient-independent seizure detector is essential. However, developing a patient-independent seizure detector is challenging as seizures exhibit diverse morphologies and characteristics across different patients and recording devices. In this study, we propose a patient-independent seizure detector to automatically detect seizures in both scalp EEG (sEEG) and intracranial EEG (iEEG). First, we deploy a convolutional neural network (CNN) with transformers (TRF) and belief matching (BM) loss to detect seizures in single-channel EEG segments (channel-level detection). Next, we extract regional features from the channel-level outputs to detect seizures in multi-channel EEG segments (segment-level detection). At last, we apply postprocessing filters to the segment-level outputs to determine the start and end points of seizures in multi-channel EEGs (EEG-level detection). We introduce the minimum overlap evaluation scoring (MOES) as an evaluation metric that accounts for minimum overlap between the detection and seizure, improving upon existing assessment metrics. We trained the seizure detector on the Temple University Hospital Seizure (TUH-SZ) sEEG dataset and evaluated it on five other independent sEEG and iEEG datasets. On the TUH-SZ dataset, the proposed patient-independent seizure detector achieves a sensitivity (SEN), precision (PRE), average and median false positive rate per hour (aFPR/h and mFPR/h), and median offset of 0.772, 0.429, 4.425, 0, and -2.125s, respectively. Across all four adult datasets (excluding neonatal and paediatric datasets), we obtained SEN of 0.617-1.00, PRE of 0.534-1.00, aFPR/h of 0.425-2.002, and mFPR/h of 0-1.003. Meanwhile, on neonatal and paediatric datasets, we obtained SEN of 0.227-0.678, PRE of 0.377-0.818, aFPR/h of 0.253-0.421, and mFPR/h of 0.118-0.223. The proposed seizure detector can reliably detect seizures in adult EEGs (to less extent in neonatal EEGs) and takes less than 15s for a 30 minutes EEG. Hence, this system could potentially aid the clinicians in reliably identifying seizures expeditiously, allocating more time for devising proper treatment.

show abstract

Section: Benefits Of Channel-level Seizure Detectionsupporting

confidence: 91%

Section: Introductionmentioning

confidence: 99%

Multi-center Assessment of CNN-Transformer with Belief Matching Loss for Patient-independent Seizure Detection in Scalp and Intracranial EEG

Peh

Thangavel

Yao

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The literature also demonstrates the time-delay neural network for classification purposes. Thyagachandran et al ( 2020 ) used this approach to classify EEG signals; however, the presented model was not sufficiently deep to learn the hierarchical features of the EEG signal. The research that resonated most with our present study is that of Tanveer et al ( 2019 ).…”

Section: Introductionmentioning

confidence: 99%

EEG-fNIRS-based hybrid image construction and classification using CNN-LSTM

et al. 2022

View full text Add to dashboard Cite

The constantly evolving human–machine interaction and advancement in sociotechnical systems have made it essential to analyze vital human factors such as mental workload, vigilance, fatigue, and stress by monitoring brain states for optimum performance and human safety. Similarly, brain signals have become paramount for rehabilitation and assistive purposes in fields such as brain–computer interface (BCI) and closed-loop neuromodulation for neurological disorders and motor disabilities. The complexity, non-stationary nature, and low signal-to-noise ratio of brain signals pose significant challenges for researchers to design robust and reliable BCI systems to accurately detect meaningful changes in brain states outside the laboratory environment. Different neuroimaging modalities are used in hybrid settings to enhance accuracy, increase control commands, and decrease the time required for brain activity detection. Functional near-infrared spectroscopy (fNIRS) and electroencephalography (EEG) measure the hemodynamic and electrical activity of the brain with a good spatial and temporal resolution, respectively. However, in hybrid settings, where both modalities enhance the output performance of BCI, their data compatibility due to the huge discrepancy between their sampling rate and the number of channels remains a challenge for real-time BCI applications. Traditional methods, such as downsampling and channel selection, result in important information loss while making both modalities compatible. In this study, we present a novel recurrence plot (RP)-based time-distributed convolutional neural network and long short-term memory (CNN-LSTM) algorithm for the integrated classification of fNIRS EEG for hybrid BCI applications. The acquired brain signals are first projected into a non-linear dimension with RPs and fed into the CNN to extract essential features without performing any downsampling. Then, LSTM is used to learn the chronological features and time-dependence relation to detect brain activity. The average accuracies achieved with the proposed model were 78.44% for fNIRS, 86.24% for EEG, and 88.41% for hybrid EEG-fNIRS BCI. Moreover, the maximum accuracies achieved were 85.9, 88.1, and 92.4%, respectively. The results confirm the viability of the RP-based deep-learning algorithm for successful BCI systems.

show abstract

“…However, there are also many that deployed their own private seizure datasets. Many applied different variations of machine learning models [297], convolutional neural networks (CNN) [216,290,291,294,296], recurrent neural networks (RNN) [291,293], long short-term memory (LSTM) [298,299], transformer [228,229], transfer learning models [293,[300][301][302], and temporal graph convolutional networks (TGCN) [303].…”

Section: Automated Seizure Detectorsmentioning

confidence: 99%

“…In recent studies on automated seizure detection from EEG, the seizure detectors are validated mainly on two public seizure datasets: the Temple University Hospital seizure (TUH-SZ) dataset [291][292][293][294] and the Children's Hospital Boston Massachusetts Institute of Technology (CHB-MIT) dataset [290,292,295,296]. In many studies, different seizure detectors are proposed, including (standard) machine learning models [297,343], convolutional neural networks (CNNs) [216,290,291,294,296], recurrent neural networks (RNNs) [291,293], long short-term memory (LSTM) [298,299], transformer [228,229], transfer learning models [293,[300][301][302], and temporal graph convolutional networks (TGCNs) [303]. The seizure detectors proposed in these studies are similar in architecture or implementation.…”

Section: Patient-independent Seizure Detection In Eeg and Ieegmentioning

confidence: 99%

Deep learning methods for diagnosis of epilepsy from electroencephalograms

Peh¹

View full text Add to dashboard Cite

Seizure Detection Using Time Delay Neural Networks and LSTMs

Cited by 7 publications

References 17 publications

Multi-center Assessment of CNN-Transformer with Belief Matching Loss for Patient-independent Seizure Detection in Scalp and Intracranial EEG

Multi-center Assessment of CNN-Transformer with Belief Matching Loss for Patient-independent Seizure Detection in Scalp and Intracranial EEG

EEG-fNIRS-based hybrid image construction and classification using CNN-LSTM

Deep learning methods for diagnosis of epilepsy from electroencephalograms

Contact Info

Product

Resources

About