Personalizing Heart Rate-Based Seizure Detection Using Supervised SVM Transfer Learning

Abstract

Objective: Automated seizure detection is a key aspect of wearable seizure warning systems. As a result, the quality of life of refractory epilepsy patients could be improved. Most state-of-the-art algorithms for heart rate-based seizure detection use a so-called patient-independent approach, which do not take into account patient-specific data during algorithm training. Although such systems are easy to use in practice, they lead to many false detections as the ictal heart rate changes are patient-dependent. In practice, only a limited amount of accurately annotated patient data is typically available, which makes it difficult to create fully patient-specific algorithms. Methods: In this context, this study proposes for the first time a new transfer learning approach that allows to personalize heart rate-based seizure detection by using only a couple of days of data per patient. The algorithm was evaluated on 2,172 h of single-lead ECG data from 24 temporal lobe epilepsy patients including 227 focal impaired awareness seizures. Results: The proposed personalized approach resulted in an overall sensitivity of 71% with 1.9 false detections per hour. This is an average decrease in false detection rate of 37% compared to the reference patient-independent algorithm using only a limited amount of personal seizure data. The proposed transfer learning approach adapts faster and more robustly to patient-specific characteristics than other alternatives for personalization in the literature. Conclusion: The proposed method allows an easy implementable solution to personalize heart rate-based seizure detection, which can improve the quality of life of refractory epilepsy patients when used as part of a multimodal seizure detection system.

Supervised SVM Transfer Learning for Modality-Specific Artefact Detection in ECG

“…Support vector machines (SVM) classification models with an RBF Kernel were used as base classifiers for the transfer learning. These classifiers are adapted using the transfer learning formulation described in [ 15 ].…”

“…We optimised each base classifier for the other datasets by applying the transfer learning approach described in [ 15 ]. The key concept of this approach is the modification of the objective function of the SVM.…”

“…Let be the M training points of the new dataset, with as the input data, and , the corresponding labels. De Cooman et al proposed the following optimisation problem for an adapted SVM classifier [ 15 ], with the weight vector of the adapted classifier, D the regularisation constant, similarly to C in Equation ( 1 ), an additional weight constant and the error of the model on data point . The factor was used in [ 15 ] to counter the class imbalance in the dataset.…”

“…De Cooman et al proposed the following optimisation problem for an adapted SVM classifier [ 15 ], with the weight vector of the adapted classifier, D the regularisation constant, similarly to C in Equation ( 1 ), an additional weight constant and the error of the model on data point . The factor was used in [ 15 ] to counter the class imbalance in the dataset. In this study, we always selected an equal amount of samples from both classes, so that the subsets for transfer learning were always balanced.…”

“…The second goal was to optimise the performance of this artefact detection model towards ccECG. To this end, we made use of the transfer learning formulation described in [ 15 ], to create a modality-specific classifier by using only a limited amount of modality-specific data. Last, we investigated whether an active sampling strategy based on entropy could decrease the amount of samples that are needed to optimise the model.…”