Detecting Tonic-Clonic Seizures in Multimodal Biosignal Data From Wearables: Methodology Design and Validation

Abstract

Background Video electroencephalography recordings, routinely used in epilepsy monitoring units, are the gold standard for monitoring epileptic seizures. However, monitoring is also needed in the day-to-day lives of people with epilepsy, where video electroencephalography is not feasible. Wearables could fill this gap by providing patients with an accurate log of their seizures. Objective Although there are already systems available that provide promising results for the detection of tonic-clonic seizures (TCSs), research in this area is often limited to detection from 1 biosignal modality or only during the night when the patient is in bed. The aim of this study is to provide evidence that supervised machine learning can detect TCSs from multimodal data in a new data set during daytime and nighttime. Methods An extensive data set of biosignals from a multimodal watch worn by people with epilepsy was recorded during their stay in the epilepsy monitoring unit at 2 European clinical sites. From a larger data set of 243 enrolled participants, those who had data recorded during TCSs were selected, amounting to 10 participants with 21 TCSs. Accelerometry and electrodermal activity recorded by the wearable device were used for analysis, and seizure manifestation was annotated in detail by clinical experts. Ten accelerometry and 3 electrodermal activity features were calculated for sliding windows of variable size across the data. A gradient tree boosting algorithm was used for seizure detection, and the optimal parameter combination was determined in a leave-one-participant-out cross-validation on a training set of 10 seizures from 8 participants. The model was then evaluated on an out-of-sample test set of 11 seizures from the remaining 2 participants. To assess specificity, we additionally analyzed data from up to 29 participants without TCSs during the model evaluation. Results In the leave-one-participant-out cross-validation, the model optimized for sensitivity could detect all 10 seizures with a false alarm rate of 0.46 per day in 17.3 days of data. In a test set of 11 out-of-sample TCSs, amounting to 8.3 days of data, the model could detect 10 seizures and produced no false positives. Increasing the test set to include data from 28 more participants without additional TCSs resulted in a false alarm rate of 0.19 per day in 78 days of wearable data. Conclusions We show that a gradient tree boosting machine can robustly detect TCSs from multimodal wearable data in an original data set and that even with very limited training data, supervised machine learning can achieve a high sensitivity and low false-positive rate. This methodology may offer a promising way to approach wearable-based nonconvulsive seizure detection.

Machine learning model evaluation on healthcare applications: A Review

“…Some articles reported tests carried out in real-world environments with patients. However, they didn't detail their evaluation procedures on production models (Lam et al, 2022;Birkenbihl et al, 2020;Kamran et al, 2022;The RADAR-CNS Consortium et al, 2021). It's also noticeable that there is little information about the metrics and best practices for model evaluation in production for healthcare applications.…”

Seizure detection based on wearable devices: A review of device, mechanism, and algorithm

“…Based on algorithms such as SVM or LSTM, the sensitivity of 91%-94.55% and the FDR of 0.01-0.1/h were obtained. 18,19,62,68,69 ECG and behind-the-ear EEG play an important role in the detection of seizures that are not limited to motor types, but the results were worse than those just for motor seizures detection, with sensitivity ranging from 75% to 92% and FDR ranging from 0.56 to 1.85/h. 63,70 Epileptic seizure detection based on single-modal signal still has good research prospects due to the availability of data, portability of instruments, and ease of commercialization.…”

“…When detecting motor seizures, ACM, sEMG, EDA, ECG, and skin temperature have been used. Based on algorithms such as SVM or LSTM, the sensitivity of 91%–94.55% and the FDR of 0.01–0.1/h were obtained 18,19,62,68,69 . ECG and behind‐the‐ear EEG play an important role in the detection of seizures that are not limited to motor types, but the results were worse than those just for motor seizures detection, with sensitivity ranging from 75% to 92% and FDR ranging from 0.56 to 1.85/h 63,70 …”

Feasibility of cardiac‐based seizure detection and prediction: A systematic review of non‐invasive wearable sensor‐based studies

“…11,12 Research on seizure detection or prediction based on non-cerebral signals has grown significantly due to the rising prevalence of wearable devices that can non-invasively measure signals such as accelerometer (ACC), electrocardiogram (ECG), electrodermal activity (EDA), and electromyography. [13][14][15][16] By coupling these measurements with the application of machine learning tools, substantial progress has been made in generating new insights into seizure patterns (Figure 1). Dysfunction in the autonomic nervous system (ANS) has been particularly observed in focal seizures with a temporal lobe origin, as well as focal-to-bilateral and generalized tonic-clonic seizures.…”