A Self-Aware Epilepsy Monitoring System for Real-Time Epileptic Seizure Detection

Forooghifar, Farnaz; Aminifar, Amir; Cammoun, Leila; Wisniewski, Ilona; Ciumas, Carolina; Ryvlin, Philippe; Atienza, David

doi:10.1007/s11036-019-01322-7

Cited by 37 publications

(36 citation statements)

References 72 publications

(74 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we choose Random Forest (RF) as the classification algorithm as it is lightweight, suitable for resourceconstrained platforms, which has already been shown to provide promising results for epilepsy detection [21]. The the strength of individual trees and their diversity contributes to the performance of a RF ensemble of classification and regression trees.…”

Section: Seizure Classificationmentioning

confidence: 99%

Robust Epileptic Seizure Detection on Wearable Systems with Reduced False-Alarm Rate

Zanetti

Aminifar

Atienza

2020

2020 42nd Annual International Conference of the IEEE Engineering in Medicine &Amp; Biology Society (EMBC)

Self Cite

View full text Add to dashboard Cite

Epilepsy affects more than 50 million people and ranks among the most common neurological diseases worldwide. Despite advances in treatment, one-third of patients still suffer from refractory epilepsy. Wearable devices for real-time patient monitoring can potentially improve the quality of life for such patients and reduce the mortality rate due to seizure-related accidents and sudden death in epilepsy. However, the majority of employed seizure detection techniques and devices suffer from unacceptable false-alarm rate. In this paper, we propose a robust seizure detection methodology for a wearable platform and validate it on the Physionet.org CHB-MIT Scalp EEG database. It reaches sensitivity of 0.966 and specificity of 0.925, and reducing the false-alarm rate by 34.7%. We also evaluate the battery lifetime of the wearable system including our proposed methodology and demonstrate the feasibility of using it in real time for up to 40.87 hours on a single battery charge. Clinical relevance-We propose a methodology to increase classification robustness and reduce the false-alarm rate for epileptic seizure detection using wearable systems.

show abstract

Section: Seizure Classificationmentioning

confidence: 99%

Robust Epileptic Seizure Detection on Wearable Systems with Reduced False-Alarm Rate

Zanetti

Aminifar

Atienza

2020

2020 42nd Annual International Conference of the IEEE Engineering in Medicine &Amp; Biology Society (EMBC)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Wearable sensors allow monitoring specific characteristics of a patient's pathology by measuring bio-signals with non-invasive sensors, e.g., electrocardiogram (ECG) and electroencephalogram (EEG). Therefore, they can enable accurate detection and prediction of major noncommunicable diseases (NCDs) and allow people to self-assess of their health status [2]- [7].…”

Section: Introductionmentioning

confidence: 99%

“…Despite recent advances in wearable technologies, major challenges exist to fully exploit such systems. In particular, energy efficiency and scalability (i.e., according to the specific pathology characteristics of each patient) are important factors to take into account in any wearable sensor design [8] for personalized remote long-term health monitoring [5]- [7], [9]- [11]. Modern ultra-low power (ULP) platforms [12]- [16] can offer many advantages in terms of parallelization capabilities that can be exploited in biomedical applications.…”

Section: Introductionmentioning

confidence: 99%

Real-Time Personalized Atrial Fibrillation Prediction on Multi-Core Wearable Sensors

Giovanni

Valdés

Peon

et al. 2021

IEEE Trans. Emerg. Topics Comput.

Self Cite

View full text Add to dashboard Cite

In the recent Internet-of-Things (IoT) era where biomedical applications require continuous monitoring of relevant data, edge computing keeps gaining more and more importance. These new architectures for edge computing include multi-core and parallel computing capabilities that can enable prevention diagnosis and treatment of diseases in ambulatory or home-based setups. In this paper, we explore the benefits of the parallelization capabilities and computing heterogeneity of new wearable sensors in the context of a personalized online atrial fibrillation (AF) prediction method for daily monitoring. First, we apply optimizations to a single-core design to reduce energy, based on patient-specific training models. Second, we explore multi-core and memory banks configuration changes to adapt the computation and storage requirements to the characteristics of each patient. We evaluate our methodology on the Physionet Prediction Challenge (2001) publicly available database, and assess the energy consumption of single-core (ARM Cortex-M3 based) and new ultra-low power multi-core architectures (opensource RISC-V based) for next-generation of wearable platforms. Overall, our exploration at the application level highlights that a parallelization approach for personalized AF in multi-core wearable sensors enables energy savings up to 24 % with respect to single-core sensors. Moreover, including the adaptation of the memory subsystem (size and number of memory banks), in combination with deep sleep energy saving modes, can overall provide total energy savings up to 34 %, depending on the specific patient.

show abstract

“…In this way, WSNs have evolved from single-core systems [6], [7] into ultra-low power (ULP) [8] and multi-core parallel computing platforms [9]- [13]. Most of the typical WSN-based biomedical applications in the state-of-the-art have been implemented on single-core processors [6], [7], [14], [15]. To exploit the new parallel capabilities of modern WSN platforms in the context of biomedical applications, per-lead (i.e.…”

mentioning

confidence: 99%

“…channel) multi-core computation is a natural option to achieve low-power operation, as in the case of multi-lead ECG analysis [10], [16]. However, more general WSN-based biomedical applications for monitoring of NCDs typically include several building blocks which often are not amenable to per-lead parallelization [7], [11], [14], [15], [17]- [22]. Modern platforms have also evolved into hybrid systems with a main core and an additional cluster of cores [9] that allow flexible design of efficient single-core and parallel modules, in applications where several modules cannot be parallelized easily.…”

mentioning

confidence: 99%

Modular Design and Optimization of Biomedical Applications for Ultralow Power Heterogeneous Platforms

Giovanni

Montagna

Denkinger

et al. 2020

IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.

Self Cite

View full text Add to dashboard Cite

In the last years, remote health monitoring is becoming an essential branch of health care with the rapid development of wearable sensors technology. To meet the demand of new more complex applications and ensuring adequate battery lifetime, wearable sensors have evolved into multi-core systems with advanced power-saving capabilities and additional heterogeneous components. In this paper, we present an approach that applies optimization and parallelization techniques uncovered by modern ultra-low power platforms in the SW layers with the goal of improving the mapping and reducing the energy consumption of biomedical applications. Additionally, we investigate the benefit of integrating domain-specific accelerators to further reduce the energy consumption of the most computationally expensive kernels. Using 30-second excerpts of signals from two public databases, we apply the proposed optimization techniques on well-known modules of biomedical benchmarks from the stateof-the-art and two complete applications. We observe speedups of 5.17 × and energy savings of 41.6 % for the multicore implementation using a cluster of 8 cores with respect to single-core wearable sensor designs when processing a standard 12-lead electrocardiogram (ECG) signal analysis. Additionally, we conclude that the minimum workload required to take advantage of parallelization for a hearbeat classifier corresponds to the processing of 3-lead ECG signals, with a speed-up of 2.96 × and energy savings of 19.3 %. Moreover, we observe additional energy savings of up to 7.75 % and 16.8 % by applying power management and memory scaling to the multicore implementation of the 3-lead beat classifier and 12-lead ECG analysis, respectively. Finally, by integrating hardware (HW) acceleration we observe overall energy savings of up to 51.3 % for the 12-lead ECG analysis. I. INTRODUCTION I NCREASING healthcare costs [1] and hospital overcrowding call for new technological advances that improve re

show abstract

A Self-Aware Epilepsy Monitoring System for Real-Time Epileptic Seizure Detection

Cited by 37 publications

References 72 publications

Robust Epileptic Seizure Detection on Wearable Systems with Reduced False-Alarm Rate

Robust Epileptic Seizure Detection on Wearable Systems with Reduced False-Alarm Rate

Real-Time Personalized Atrial Fibrillation Prediction on Multi-Core Wearable Sensors

Modular Design and Optimization of Biomedical Applications for Ultralow Power Heterogeneous Platforms

Contact Info

Product

Resources

About