Automatic Detection and Classification of Epileptic Seizures from EEG Data: Finding Optimal Acquisition Settings and Testing Interpretable Machine Learning Approach

被引:8
作者
Statsenko, Yauhen [1 ,2 ,3 ]
Babushkin, Vladimir [1 ]
Talako, Tatsiana [1 ,4 ]
Kurbatova, Tetiana [1 ]
Smetanina, Darya [1 ]
Simiyu, Gillian Lylian [1 ]
Habuza, Tetiana [3 ,5 ]
Ismail, Fatima [6 ]
Almansoori, Taleb M. [1 ]
Gorkom, Klaus N. -V. [1 ]
Szolics, Miklos [7 ,8 ]
Hassan, Ali [7 ]
Ljubisavljevic, Milos [9 ,10 ]
机构
[1] United Arab Emirates Univ, Coll Med & Hlth Sci, Radiol Dept, POB 15551, Al Ain, U Arab Emirates
[2] ASPIRE Precis Med Res Inst Abu Dhabi, Med Imaging Platform, POB 15551, Al Ain, U Arab Emirates
[3] United Arab Emirates Univ, Big Data Analyt Ctr, POB 15551, Al Ain, U Arab Emirates
[4] Minsk Sci & Pract Ctr Surg Transplantol & Hematol, Dept Oncohematol, Minsk 220089, BELARUS
[5] United Arab Emirates Univ, Coll Informat Technol, Dept Comp Sci & Software Engn, POB 15551, Al Ain, U Arab Emirates
[6] United Arab Emirates Univ, Coll Med & Hlth Sci, Pediat Dept, POB 15551, Al Ain, U Arab Emirates
[7] Tawam Hosp, Med Dept, Neurol Div, POB 15258, Al Ain, U Arab Emirates
[8] United Arab Emirates Univ, Coll Med & Hlth Sci, Internal Med Dept, POB 15551, Al Ain, U Arab Emirates
[9] United Arab Emirates Univ, Dept Physiol, Coll Med & Hlth Sci, POB 15551, Al Ain, U Arab Emirates
[10] ASPIRE Precis Med Res Inst Abu Dhabi, Neurosci Platform, POB 15551, Al Ain, U Arab Emirates
关键词
deep learning; interpretable machine learning; activation maximization; epileptic seizure; EEG; acquisition settings; source reconstruction; HIGH-FREQUENCY OSCILLATIONS; FEATURE-EXTRACTION; ABSENCE SEIZURES; NEURAL-NETWORKS; SYSTEM; BRAIN; MISDIAGNOSIS; CONNECTIONS; TRANSFORM; PATTERNS;
D O I
10.3390/biomedicines11092370
中图分类号
Q5 [生物化学]; Q7 [分子生物学];
学科分类号
071010 ; 081704 ;
摘要
Deep learning (DL) is emerging as a successful technique for automatic detection and differentiation of spontaneous seizures that may otherwise be missed or misclassified. Herein, we propose a system architecture based on top-performing DL models for binary and multigroup classifications with the non-overlapping window technique, which we tested on the TUSZ dataset. The system accurately detects seizure episodes (87.7% Sn, 91.16% Sp) and carefully distinguishes eight seizure types (95-100% Acc). An increase in EEG sampling rate from 50 to 250 Hz boosted model performance: the precision of seizure detection rose by 5%, and seizure differentiation by 7%. A low sampling rate is a reasonable solution for training reliable models with EEG data. Decreasing the number of EEG electrodes from 21 to 8 did not affect seizure detection but worsened seizure differentiation significantly: 98.24 +/- 0.17 vs. 85.14 +/- 3.14% recall. In detecting epileptic episodes, all electrodes provided equally informative input, but in seizure differentiation, their informative value varied. We improved model explainability with interpretable ML. Activation maximization highlighted the presence of EEG patterns specific to eight seizure types. Cortical projection of epileptic sources depicted differences between generalized and focal seizures. Interpretable ML techniques confirmed that our system recognizes biologically meaningful features as indicators of epileptic activity in EEG.
引用
收藏
页数:22
相关论文
共 50 条
  • [31] From Sound Perception to Automatic Detection of Schizophrenia: An EEG-Based Deep Learning Approach
    Barros, Carla
    Roach, Brian
    Ford, Judith M.
    Pinheiro, Ana P.
    Silva, Carlos A.
    FRONTIERS IN PSYCHIATRY, 2022, 12
  • [32] DETECTION OF EPILEPTIC CONVULSIONS FROM ACCELEROMETRY SIGNALS THROUGH MACHINE LEARNING APPROACH
    Milosevic, Milica
    Van de Vel, Anouk
    Bonroy, Bert
    Ceulemans, Berten
    Lagae, Lieven
    Vanrumste, Bart
    Van Huffel, Sabine
    2014 IEEE INTERNATIONAL WORKSHOP ON MACHINE LEARNING FOR SIGNAL PROCESSING (MLSP), 2014,
  • [33] A novel end-to-end approach for epileptic seizure classification from scalp EEG data using deep learning technique
    Kumar P.R.
    Shilpa B.
    Jha R.K.
    Mohanty S.N.
    International Journal of Information Technology, 2023, 15 (8) : 4223 - 4231
  • [34] EEG epileptic seizure detection and classification based on dual-tree complex wavelet transform and machine learning algorithms
    Ben Slimen, Itaf
    Boubchir, Larbi
    Mbarki, Zouhair
    Seddik, Hassene
    JOURNAL OF BIOMEDICAL RESEARCH, 2020, 34 (03): : 151 - 161
  • [35] EEG epileptic seizure detection and classification based on dual-tree complex wavelet transform and machine learning algorithms
    Itaf Ben Slimen
    Larbi Boubchir
    Zouhair Mbarki
    Hassene Seddik
    The Journal of Biomedical Research, 2020, 34 (03) : 151 - 161
  • [36] A Fully Automatic, Interpretable and Adaptive Machine Learning Approach to Map Burned Area from Remote Sensing
    Stroppiana, Daniela
    Bordogna, Gloria
    Sali, Matteo
    Boschetti, Mirco
    Sona, Giovanna
    Brivio, Pietro Alessandro
    ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION, 2021, 10 (08)
  • [37] Cloud based ensemble machine learning approach for smart detection of epileptic seizures using higher order spectral analysis
    Kuldeep Singh
    Jyoteesh Malhotra
    Physical and Engineering Sciences in Medicine, 2021, 44 : 313 - 324
  • [38] Automatic large-scale data acquisition via crowdsourcing for crosswalk classification: A deep learning approach
    Berriel, Rodrigo F.
    Rossi, Franco Schmidt
    de Souza, Alberto F.
    Oliveira-Santos, Thiago
    COMPUTERS & GRAPHICS-UK, 2017, 68 : 32 - 42
  • [39] EPILEPTIC EEG SIGNALS RHYTHMS ANALYSIS IN THE DETECTION OF FOCAL AND NON-FOCAL SEIZURES BASED ON OPTIMISED MACHINE LEARNING AND DEEP NEURAL NETWORK ARCHITECTURE
    Saminu, Sani
    Xu, Guizhi
    Shuai, Zhang
    El Kader, Isselmou abd
    Jabire, Adamu halilu
    Ahmed, Yusuf kola
    Karaye, Ibrahim abdullahi
    Ahmad, Isah salim
    JOURNAL OF MECHANICS IN MEDICINE AND BIOLOGY, 2024, 24 (01)
  • [40] Early Detection of Alcohol Use Disorder Based on a Novel Machine Learning Approach Using EEG Data
    Flathau, Dennis
    Breitenbach, Johannes
    Baumgartl, Hermann
    Buettner, Ricardo
    2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), 2021, : 3897 - 3904