Towards an Online Seizure Advisory System—An Adaptive Seizure Prediction Framework Using Active Learning Heuristics

Abstract

In the last decade, seizure prediction systems have gained a lot of attention because of their enormous potential to largely improve the quality-of-life of the epileptic patients. The accuracy of the prediction algorithms to detect seizure in real-world applications is largely limited because the brain signals are inherently uncertain and affected by various factors, such as environment, age, drug intake, etc., in addition to the internal artefacts that occur during the process of recording the brain signals. To deal with such ambiguity, researchers transitionally use active learning, which selects the ambiguous data to be annotated by an expert and updates the classification model dynamically. However, selecting the particular data from a pool of large ambiguous datasets to be labelled by an expert is still a challenging problem. In this paper, we propose an active learning-based prediction framework that aims to improve the accuracy of the prediction with a minimum number of labelled data. The core technique of our framework is employing the Bernoulli-Gaussian Mixture model (BGMM) to determine the feature samples that have the most ambiguity to be annotated by an expert. By doing so, our approach facilitates expert intervention as well as increasing medical reliability. We evaluate seven different classifiers in terms of the classification time and memory required. An active learning framework built on top of the best performing classifier is evaluated in terms of required annotation effort to achieve a high level of prediction accuracy. The results show that our approach can achieve the same accuracy as a Support Vector Machine (SVM) classifier using only 20% of the labelled data and also improve the prediction accuracy even under the noisy condition.

Pattern Recognition in Epileptic EEG Signals via Dynamic Mode Decomposition

“…However, this approach is not scalable to large patient groups as developing patient-specific machine learning models requires enormous efforts in collecting, labeling, and training the machine learning algorithms. A number of studies are underway on the development of an online seizure advisory system that automates this process [35]. However, due to the risk and complexity involved in ECoG, it is strictly restricted to clinical use.…”

Prediction of Seizure Recurrence. A Note of Caution

“…Despite subsequent expressions of enthusiasm ( 12 , 31 , 92 , 127 , 128 ), others have found that the EEGs of one third of patients with focal ( 129 ) or multifocal ( 130 ) epilepsies were not able to provide adequate predictive information about impending seizures. Findings such as these prompt us to offer words of caution about the anticipated capability to predict seizures and intervene effectively to prevent seizure occurrence.…”

Adaptive, Unlabeled and Real-time Approximate-Learning Platform (AURA) for Personalized Epileptic Seizure Forecasting

“…These system models that can still be identified as cloud/expert-in-the-loop or phone/patient-in-the-loop, require continuous or, if pre-identified, regular data telemetry for their machine learning models to be trained on the cloud or smartphone. Many works can be identically or partially identified as offloading resource intensive tasks externally, with reliance on distributed software at the heart of these closed-loop systems [22,23,24,25,26,27,28,29,30,31,32]. To list a few examples, the role of a processing unit, in innovations such as [32], is limited to the controller of system operation.…”

“…These systems rely on either continuous or regular data telemetry of the data stored in the database. Examples include [22,23,24,25,26,27,28,29,30,31,32]. (e) AURA operates on the basis of a neuromorphic neuromodulation system, where the medical device hosts online training and active learning without any reliance on external computing resources.…”