Seizure-Detection

Date: 9/11/2023

Dataset Introduction

Source: CHB-MIT Scalp EEG Database; Link: https://physionet.org/content/chbmit/1.0.0/

Code Introduction

seizure_detection_phase_1:

  • Read data(chb07_13)
  • PCA method to reduce raw signal dimensionality
  • Features calculation: Mean, variance, Mode, Median, Peaks, Signal energy(Time Domain), Approximate entropy(Not results generated), Spectrum energy, Continuous wavelet transform(CWT)

Paper_parallel_memristive_convolution_network:
This code is based on the paper called Seizure_Detection_and_Prediction_by_Parallel_Memristive_Convolutional_Neural_Networks and author: Chenqi Li

  • Feature selection: Mean, Variance, Skewness, Kurtosis, Coefficient of Variation, Median absolute deviation, Root mean square amplitude, Shannon entropy.
    Continue working in this code...

Data 9/24/2023

  • I separated Paper_parallel_memristive_convolution_network into 'FeatureSelection&ArraySave' and 'Network_Train_Evaluation'.
    • FeatureSelection: Calculating 8 features, using PCA reduce feature dimension into 64, Save (# sample, # feature) arrays.
    • Network_Train_Evaluation: Applying saved metrix to train model created by paper's author and generate evaluation metrix.

Data 10/23/2023

  • Add Data folder
    • Include seizure and non-seizure data for patient 01, 02, 07.
  • Preprocessing
    • "Patient_Nonseizure.ipynb" calculating non-seizure period features.
    • "patient_Seizure.ipynb" calculating siezure period(40min before the start of seizure) features.
  • Update
    • "Network_Train_Evaluatin.ipynb" adding confusion matrix, classification report.
    • "Functions.py" has all feature extraction methods.

Reference

  • Siddiqui, M. K., Morales-Menendez, R., Huang, X., & Hussain, N. (2020). A review of epileptic seizure detection using machine learning classifiers. Brain informatics, 7(1), 1-18.
  • Li, C., Lammie, C., Amirsoleimani, A., Azghadi, M. R., & Genov, R. (2023). Simulation of memristive crossbar arrays for seizure detection and prediction using parallel Convolutional Neural Networks. Software Impacts, 15, 100473. Source code

Dependency

  • numpy
  • statistics
  • pyedflib
  • sklearn
  • matplotlib
  • IPython