EEGformer: Transformer-Based Epilepsy Detection on Raw EEG Traces for Low-Channel-Count Wearable Continuous Monitoring Devices

Busia, Paola; Cossettini, Andrea; Ingolfsson, Thorir Mar; Benatti, Simone; Burrello, Alessio; Scherer, Moritz; Scrugli, Matteo Antonio; Meloni, Paolo; Benini, Luca

doi:10.1109/biocas54905.2022.9948637

Cited by 10 publications

(9 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors of [13] do not report on the specificity of their network but reproduced numbers showcase that it achieves a very high specificity (99.75% -99.86%). A noticeable decrease in sensitivity from (81.91% → 68.73%) aligns with the figures documented in [14], which similarly employs a four-channel temporal framework. It merits highlighting that the authors of [14] incorporated only a subset of the total 23 patients (specifically 8), hence constraining a holistic comparison of sensitivity and specificity metrics.…”

Section: A Chb-mitsupporting

confidence: 83%

“…Although their CNN architecture achieved a reported sensitivity of 99.81%, it was not validated on an embedded device or designed to accommodate a lower channel count. In contrast, [14] represents the current benchmark, addressing the challenges associated with reduced electrode montage for wearable technologies and achieving 99.9% specificity, 65.5% sensitivity, and 0.8 FP/h on the CHB-MIT dataset. However, this network was only validated on a subset of the CHB-MIT dataset, obscuring its generalizability across all patients or diverse datasets.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

EpiDeNet: An Energy-Efficient Approach to Seizure Detection for Embedded Systems

Ingolfsson,

Chakraborty,

Wang

et al. 2023

2023 IEEE Biomedical Circuits and Systems Conference (BioCAS)

Self Cite

View full text Add to dashboard Cite

Section: A Chb-mitsupporting

confidence: 83%

Section: Introductionmentioning

confidence: 99%

EpiDeNet: An Energy-Efficient Approach to Seizure Detection for Embedded Systems

Ingolfsson,

Chakraborty,

Wang

et al. 2023

2023 IEEE Biomedical Circuits and Systems Conference (BioCAS)

Self Cite

View full text Add to dashboard Cite

“…We designed a TSD system to identify epilepsy seizures using pre-recorded EEG signals by short-time Fourier transform (STFT) on the most extensive publicly available EEG dataset, TUH. This paper is part of a newly formed set of papers analysing the use of Transformers on EEG signals and seizure detection to locate and detect epilepsy seizures [43,41]. Unfortunately, separate Transformer architectures are rarely adopted for signal processing due to the lack of inductive biases, which contributes to the learning process [44,45].…”

Section: B Noveltymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

TSD: Transformers for Seizure Detection

Liu

Sabrina

et al. 2023

Preprint

View full text Add to dashboard Cite

Epilepsy is a common neurological disorder that substantially deteriorates patients' safety and quality of life. Electroencephalogram (EEG) has been the golden-standard technique for diagnosing this brain disorder and has played an essential role in epilepsy monitoring and disease management. It is extremely laborious and challenging, if not practical, for physicians and expert humans to annotate all recorded signals, particularly in long-term monitoring. The annotation process often involves identifying signal segments with suspected epileptic seizure features or other abnormalities and/or known healthy features. Therefore, automated epilepsy detection becomes a key clinical need because it can greatly improve clinical practice's efficiency and free up human expert time to attend to other important tasks. Current automated seizure detection algorithms generally face two challenges: (1) models trained for specific patients, but such models are patient-specific, hence fail to generalize to other patients and real-world situations; (2) seizure detection models trained on large EEG datasets have low sensitivity and/or high false positive rates, often with an area under the receiver operating characteristic (AUROC) that is not high enough for potential clinical applicability. This paper proposes Transformers for Seizure Detection, which we refer to as TSD in this manuscript. A Transformer is a deep learning architecture based on an encoder-decoder structure and on attention mechanisms, which we apply to recorded brain signals. The AUROC of our proposed model has achieved 92.1%, tested with Temple University's publically available electroencephalogram (EEG) seizure corpus dataset (TUH). Additionally, we highlight the impact of input domains on the model's performance. Specifically, TSD performs best in identifying epileptic seizures when the input domain is a time-frequency. Finally, our proposed model for seizure detection in inference-only mode with EEG recordings shows outstanding performance in classifying seizure types and superior model initialization.

show abstract

“…Neural interfaces proved to be key components for enhancing the quality of life of disabled patients. Their range of applications stretches from hand movement decoding in patients affected by tetraplegia [1], seizure detection in pediatric subjects with intractable seizures [2], hand prosthesis control with sensory flow restoration in transradial amputees [3], speech decoding in patients with motor speech disorders [4], etc. Neural interfaces acquire neural signals, interpret the patient's intention, and use this information to accommodate the patient's request.…”

Section: Introductionmentioning

confidence: 99%

On-FPGA Spiking Neural Networks for End-to-End Neural Decoding

2023

View full text Add to dashboard Cite

In the last decades, deep learning neural decoding algorithms have gained momentum in the field of neural interfaces and neural processing systems. However, to be deployed on low-budget portable devices while maintaining real-time operability, these models must withstand strict computational and power limitations. This work presents a spike decoding system implemented on a low-end Zynq-7010 FPGA, which includes a multiplier-less spike detection pipeline and a spiking-neural-network-based decoder mapped in the programmable logic. We tested the system on two publicly available datasets and achieved comparable results with state-of-the-art neural decoders that use more complex deep learning models. The system required 7.36 times fewer parameters than the smallest architecture tested on the same dataset. Moreover, by exploiting the spike sparsity property of the neural signal, the total amount of computations is reduced by about 90% during a test carried out on real recorded data. The low computational complexity of the chosen spike detection setup, combined with the power efficiency of spiking neural networks, makes this prototype a well-suited choice for low-power real-time neural decoding at the edge.INDEX TERMS Neural decoding, spike detection, spiking neural network, FPGA, real-time, low-power.

show abstract

EEGformer: Transformer-Based Epilepsy Detection on Raw EEG Traces for Low-Channel-Count Wearable Continuous Monitoring Devices

Cited by 10 publications

References 25 publications

EpiDeNet: An Energy-Efficient Approach to Seizure Detection for Embedded Systems

EpiDeNet: An Energy-Efficient Approach to Seizure Detection for Embedded Systems

TSD: Transformers for Seizure Detection

On-FPGA Spiking Neural Networks for End-to-End Neural Decoding

Contact Info

Product

Resources

About