Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Multi-Modal Non-Invasive Deep Learning Framework for Progressive Prediction of Seizures

Ali Saeizadeh, Douglas Schonholtz, Joseph S. Neimat, Pedram Johari, Tommaso Melodia

TL;DR

This paper introduces an innovative framework designed for progressive prediction of seizures through the utilization of a Deep Learning (DL) methodology based on non-invasive multimodal sensor networks, achieving 95% sensitivity, 98% specificity, and 97% accuracy, averaged among 29 patients.

Abstract

This paper introduces an innovative framework designed for progressive (granular in time to onset) prediction of seizures through the utilization of a Deep Learning (DL) methodology based on non-invasive multi-modal sensor networks. Epilepsy, a debilitating neurological condition, affects an estimated 65 million individuals globally, with a substantial proportion facing drug-resistant epilepsy despite pharmacological interventions. To address this challenge, we advocate for predictive systems that provide timely alerts to individuals at risk, enabling them to take precautionary actions. Our framework employs advanced DL techniques and uses personalized data from a network of non-invasive electroencephalogram (EEG) and electrocardiogram (ECG) sensors, thereby enhancing prediction accuracy. The algorithms are optimized for real-time processing on edge devices, mitigating privacy concerns and minimizing data transmission overhead inherent in cloud-based solutions, ultimately preserving battery energy. Additionally, our system predicts the countdown time to seizures (with 15-minute intervals up to an hour prior to the onset), offering critical lead time for preventive actions. Our multi-modal model achieves 95% sensitivity, 98% specificity, and 97% accuracy, averaged among 29 patients.

A Multi-Modal Non-Invasive Deep Learning Framework for Progressive Prediction of Seizures

TL;DR

This paper introduces an innovative framework designed for progressive prediction of seizures through the utilization of a Deep Learning (DL) methodology based on non-invasive multimodal sensor networks, achieving 95% sensitivity, 98% specificity, and 97% accuracy, averaged among 29 patients.

Abstract

This paper introduces an innovative framework designed for progressive (granular in time to onset) prediction of seizures through the utilization of a Deep Learning (DL) methodology based on non-invasive multi-modal sensor networks. Epilepsy, a debilitating neurological condition, affects an estimated 65 million individuals globally, with a substantial proportion facing drug-resistant epilepsy despite pharmacological interventions. To address this challenge, we advocate for predictive systems that provide timely alerts to individuals at risk, enabling them to take precautionary actions. Our framework employs advanced DL techniques and uses personalized data from a network of non-invasive electroencephalogram (EEG) and electrocardiogram (ECG) sensors, thereby enhancing prediction accuracy. The algorithms are optimized for real-time processing on edge devices, mitigating privacy concerns and minimizing data transmission overhead inherent in cloud-based solutions, ultimately preserving battery energy. Additionally, our system predicts the countdown time to seizures (with 15-minute intervals up to an hour prior to the onset), offering critical lead time for preventive actions. Our multi-modal model achieves 95% sensitivity, 98% specificity, and 97% accuracy, averaged among 29 patients.

Paper Structure

This paper contains 7 sections, 4 equations, 5 figures.

Table of Contents

  1. Introduction
  2. System Overview
  3. -Based Seizure Prediction
  4. Deep Learning Model
  5. Combiner Model
  6. Experimental Results
  7. Conclusion

Figures (5)

  • Figure 1: Proposed ultrasonically connected . The gateway receives the classification results from the EEG and ECG nodes that execute deep learning algorithms and sends commands to the through its external controller using ultrasonic signals (This figure is generated by BioRender).
  • Figure 2: Deep Learning Model and Prediction System Structure
  • Figure 3: Confusion Matrices on the test dataset for different modalities. (averaged among the patients)
  • Figure 4: Average metrics among all the patients for different models.
  • Figure 5: Accuracy trend over time intervals leading up to seizure onset.