- (Oct 2019) A Review on EEG Based Epileptic Seizure Prediction Using Machine Learning Techniques
- (Oct 2019) Efficient Epileptic Seizure Prediction Based on Deep Learning
- Dataset: CHB-MIT
- Deep Convolutional Autoencoder + Bi-LSTM
- Patient specific
- (Oct 2019) Using Deep Learning and Machine Learning to Detect Epileptic Seizure with Electroencephalography (EEG) Data
- Dataset: UCI
- Training:Testing = 8:2
- Good comparison of different ML and DL models
- on arxiv -- not reviewed
- Epileptic Seizure Prediction Using Big Data and Deep Learning: Toward a Mobile System
- Dataset: Private
- (Apr 2019) Human Intracranial EEG Quantitative Analysis and Automatic Feature Learning for Epileptic Seizure Prediction
- on arxiv -- not reviewed
- (Aug 2017) Prediction of Epileptic Seizure by Analysing Time Series EEG Signal Using k-NN Classifier
- [Blog Post] (May 2019) Epileptic Seizure Classification ML Algorithms
- (March 2019) A Novel Independent RNN Approach to Classification of Seizures against Non-seizures
- Datasets: Bonn; CHB-MIT
- Useful summary of related works
- Training:Validating:Testing = 70:15:15
- Develops IndRNN method - outperforms their implementation of LSTM and CNN approaches
- Needs many more samples than traditional ML methods (long time span)
- Classification not prediction (realtime prediction using IndRNN is intended future work)
- on arxiv -- not reviewed
- (Sept 2017) Seizure Classification From EEG Signals Using Transfer Learning, Semi-Supervised Learning and TSK Fuzzy System
- Dataset: Bonn
- Introduces TSK method: not statistically better than other ML techniques, but output is more interpretable - generates "fuzzy rules" which can be written in english
IF signal in frequency band 1 is High
etc. - EEG data available: link
- Classifies whether a patient is epileptic or not (easier problem)
- (2016) Learning Robust Features using Deep Learning for Automatic Seizure Detection
- Dataset: CHB-MIT
- 3D probe positions -> 2D points -> FFT, 3 bin RGB -Cubic Interpolation-> 2D images
- (August 2015) Adversarial Representation Learning for RobustPatient-Independent Epileptic Seizure Detection
- Dataset: TUH
- Adversarial method to be patient independent (i.e. not dependent on age and gender)
- CHB-MIT
- 22 humans: 5 males 3-22, 17 females 1.5-19
- start and end of seizure is annotated
- UCI
- 500 humans
- 23.5 seconds each
- states labelled as:
- Recording of seizure activity
- They recorder the EEG from the area where the tumor was located
- Yes they identify where the region of the tumor was in the brain and recording the EEG activity from the healthy brain area
- eyes closed, means when they were recording the EEG signal the patient had their eyes closed
- eyes open, means when they were recording the EEG signal of the brain the patient had their eyes open
- Bonn
- 5 humans
- A-E sets of 100 single channel EEGs, only set E contains seizures
- Some preprocessing/filtering of data
- TUH EEG
- 316 humans
- IEEG.org
- Intracranial EEG
- Collection of many data sets
- Mostly animals (?)
- Don't know which ones are annotated
- A fuzzy classifier based detection for epileptic seizure signals
- Do Features From Short Durational Segments Classify Epileptic EEG Signals Effectively?
- Automated Epilepsy Diagnosis Using EEG With Test Set Evaluation
- A Novel Double-Index-Constrained, Multi-View, Fuzzy-Clustering Algorithm and its Application for Detecting Epilepsy Electroencephalogram Signals
- A Novel Synthetic CT Generation Method Using Multitask Maximum Entropy Clustering
- A Survey on Transfer Learning
- Collaborative Filtering for Brain-Computer Interaction Using Transfer Learning and Active Class Selection
- Domain Adaptation via Transfer Component Analysis
- Knowledge-Leverage-Based Fuzzy System and Its Modeling
- Knowledge-Leverage-Based TSK Fuzzy System Modeling
- Large margin transductive transfer learning
- Transfer learning and active transfer learning for reducing calibration data in single-trial classification of visually-evoked potentials
- Seizure Prediction by Graph Mining, Transfer Learning, and Transformation Learning
- Applying Deep Learning for Epilepsy Seizure Detection and Brain Mapping Visualization
- Optimized Deep Learning for EEG Big Data and Seizure Prediction BCI via Internet of Things
- number of correctly predicted testing data / number of total testing data
- Friedman test + Holm's post hoc test