A method for predicting neurological outcomes after cardiac arrest based on time-frequency domain feature fusion_Journal of Biomedical Engineering

Authors：

WANG Xiaojia ^1,2 , YANG Yunhan ² , WANG Shijie ² , ZHANG Yudong ² ,  GONG Yushun ³ ,  YANG Chunfeng ²

1. School of Internet of Things and Artificial Intelligence, Wuxi Vocational College of Science and Technology, Wuxi, Jiangsu 214028, P. R. China;
2. School of Computer Science and Engineering, Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications, Southeast University, Nanjing 210096, P. R. China;
3. Department of Biomedical Engineering and Imaging Medicine, Army Medical University, Chongqing 400015, P. R. China;

Corresponding?author：

GONG Yushun, Email: johnsongong@tmmu.edu.cn; YANG Chunfeng, Email: chunfeng.yang@seu.edu.cn

Keywords：

Neurological prediction; Channel selection; Time-frequency feature fusion

DOI：

10.7507/1001-5515.202603070

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

Accurate neurological outcome assessment after cardiac arrest is critical for clinical diagnosis and treatment. Existing electroencephalogram (EEG) prediction models suffer from high computational complexity, redundant full-channel data, and insufficient fusion of time-frequency domain features. This study proposes a lightweight deep learning model, WaveConv-SR-RepVGG, based on adaptive time-frequency feature fusion and reparameterized convolution. Taking the reparameterized visual convolutional network (RepVGG) as the baseline framework, the model introduces a shrinkage (SR) mechanism into the time-domain branch to suppress interference from low signal-to-noise ratio EEG signals. A multi-scale wavelet convolution (MS-Conv) module is constructed to extract multi-scale spectral features, and time-frequency features are fused to form joint representations for prognostic classification. Experiments were conducted on the PhysioNet 2023 dataset. Under the 18-channel setting, the model achieved an accuracy of 81.7%, an F1 score of 86.1%, and an AUC of 0.847. Under the 4-channel setting, the accuracy reached 81.2%, the F1 score reached 86.7%, and the AUC reached 0.877, with overall performance outperforming various mainstream deep learning algorithms. The proposed model enables neurological outcome prediction for patients after cardiac arrest and maintains stable and excellent classification performance with fewer EEG acquisition channels.

Citation： WANG Xiaojia, YANG Yunhan, WANG Shijie, ZHANG Yudong, GONG Yushun, YANG Chunfeng. A method for predicting neurological outcomes after cardiac arrest based on time-frequency domain feature fusion. Journal of Biomedical Engineering, 2026, 43(4): 711-717. doi: 10.7507/1001-5515.202603070 Copy

1.	茶興增, 張月, 張翼飛, 等. 基于兩階卷積神經網絡訓練有限心電數據的心臟驟停早期分類識別算法[J]. 生物醫學工程學雜志, 2024, 41(4): 692-699.
2.	Zheng W L, Amorim E, Jing J, et al. Predicting neurological outcome from electroencephalogram dynamics in comatose patients after cardiac arrest with deep learning[J]. IEEE Trans Biomed Eng, 2021, 69(5): 1813-1825.
3.	張唐娟, 張強, 胡悅朋, 等. 構建區域性急診體外心肺復蘇標準流程的實踐及探索[J]. 中華急診醫學雜志, 2025, 34(3): 286-291.
4.	Zhou G, Wang Y, Sun Z, et al. Survival outcome among patients with out-of-hospital cardiac arrest who received cardiopulmonary resuscitation in China: a systematic review and meta-analysis[J]. Eur J Med Res, 2023, 28(1): 8.
5.	Del Rios M, Bartos J A, Panchal A R, et al. Part 1: executive summary: 2025 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care[J]. Circulation, 2025, 152(16_suppl_2): S284-S312.
6.	Kramer P, Miera O, Berger F, et al. Prognostic value of serum biomarkers of cerebral injury in classifying neurological outcome after paediatric resuscitation[J]. Resuscitation, 2018, 122: 113-120.
7.	Scholefield B R, Tijssen J, Ganesan S L, et al. Prediction of good neurological outcome after return of circulation following paediatric cardiac arrest: a systematic review and meta-analysis[J]. Resuscitation, 2025, 207: 110483.
8.	吳雪梅, 劉薇薇, 徐莉. 基于BIS監測的鎮靜護理干預在顱腦損傷患者中的應用效果[J]. 現代醫藥衛生, 2024, 40(24): 4230-4233.
9.	Young G B. Continuous EEG monitoring in post-cardiac arrest patients: further prognostic insights[J]. Resuscitation, 2016, 109: 114.
10.	Khazanova D, Douglas V C, Amorim E. A matter of timing: EEG monitoring for neurological prognostication after cardiac arrest in the era of targeted temperature management[J]. Minerva Anestesiol, 2021, 87(6): 704-713.
11.	Aellen F M, Alnes S L, Loosli F, et al. Auditory stimulation and deep learning predict awakening from coma after cardiac arrest[J]. Brain, 2023, 146(2): 778-788.
12.	Pham S D T, Keijzer H M, Ruijter B J, et al. Outcome prediction of postanoxic coma: a comparison of automated electroencephalography analysis methods[J]. Neurocrit Care, 2022, 37(Suppl 2): 248-258.
13.	Krumm L, Kranz D D, Halimeh M, et al. Toward a universal map of EEG: a semantic, low-dimensional manifold for EEG classification, clustering, and prognostication[J]. Ann Neurol, 2025, 98(2): 357-368.
14.	Noirhomme Q, Lehembre R, Lugo Z D R, et al. Automated analysis of background EEG and reactivity during therapeutic hypothermia in comatose patients after cardiac arrest[J]. Clin EEG Neurosci, 2014, 45(1): 6-13.
15.	Zheng W L, Amorim E, Jing J, et al. Predicting neurological outcome in comatose patients after cardiac arrest with multiscale deep neural networks[J]. Resuscitation, 2021, 169: 86-94.
16.	Zabihi M, Zar A C, Grover P, et al. Hyperensemble learning from multimodal biosignals to robustly predict functional outcome after cardiac arrest[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.
17.	Kim D K, Yoon H C, Kim H S, et al. Predicting neurological outcome after cardiac arrest using a pretrained model with electroencephalography augmentation[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.
18.	Wang B, Zhang H, Yan M, et al. Predicting neurological recovery after cardiac arrest from electroencephalogram using residual network and random forest[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.
19.	Jin J, Miao Y, Daly I, et al. Correlation-based channel selection and regularized feature optimization for MI-based BCI[J]. Neural Netw, 2019, 118: 262-270.
20.	Ding X, Zhang X, Ma N, et al. RepVGG: making VGG-style ConvNets great again[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville: IEEE, 2021: 13733-13742.
21.	Zhao M, Zhong S, Fu X, et al. Deep residual shrinkage networks for fault diagnosis[J]. IEEE Trans Ind Inf, 2019, 16(7): 4681-4690.
22.	Li Y, Guo L, Liu Y, et al. A temporal-spectral-based squeeze-and-excitation feature fusion network for motor imagery EEG decoding[J]. IEEE Trans Neural Syst Rehabil Eng, 2021, 29: 1534-1545.
23.	Reyna M A, Amorim E, Sameni R, et al. Predicting neurological recovery from coma after cardiac arrest: the George B. Moody PhysioNet Challenge 2023[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.
24.	Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[PP/OL]. arXiv (2014-09-04) [2025-06-16]. https://doi.org/10.48550/arXiv.1409.1556.
25.	Donoho D L. De-noising by soft-thresholding[J]. IEEE Trans Inf Theory, 2002, 41(3): 613-627.
26.	Isogawa K, Ida T, Shiodera T, et al. Deep shrinkage convolutional neural network for adaptive noise reduction[J]. IEEE Signal Process Lett, 2017, 25(2): 224-228.
27.	Li Y, Liu Y, Cui W G, et al. Epileptic seizure detection in EEG signals using a unified temporal-spectral squeeze-and-excitation network[J]. IEEE Trans Neural Syst Rehabil Eng, 2020, 28(4): 782-794.
28.	Cong Z, Zhao M, Ling L, et al. A temporal-spectral based single-lead electroencephalogram feature fusion network may provide potential clinical biomarker for cardiac arrest[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.
29.	Herman P, Prasad G, McGinnity T M, et al. Comparative analysis of spectral approaches to feature extraction for EEG-based motor imagery classification[J]. IEEE Trans Neural Syst Rehabil Eng, 2008, 16(4): 317-326.
30.	Li Y, Wang X D, Luo M L, et al. Epileptic seizure classification of EEGs using time–frequency analysis based multiscale radial basis functions[J]. IEEE J Biomed Health Inform, 2017, 22(2): 386-397.
31.	Li Y, Cui W G, Huang H, et al. Epileptic seizure detection in EEG signals using sparse multiscale radial basis function networks and the Fisher vector approach[J]. Knowl Based Syst, 2019, 164: 96-106.
32.	Zhang Y, Liu B, Ji X, et al. Classification of EEG signals based on autoregressive model and wavelet packet decomposition[J]. Neural Process Lett, 2017, 45: 365-378.
33.	Cox D D, Savoy R L. Functional magnetic resonance imaging (fMRI) "brain reading": detecting and classifying distributed patterns of fMRI activity in human visual cortex[J]. Neuroimage, 2003, 19(2): 261-270.
34.	Li Y, Namburi P, Yu Z, et al. Voxel selection in fMRI data analysis based on sparse representation[J]. IEEE Trans Biomed Eng, 2009, 56(10): 2439-2451.
35.	Upadhyay R, Manglick A, Reddy D K, et al. Channel optimization and nonlinear feature extraction for electroencephalogram signals classification[J]. Comput Electr Eng, 2015, 45: 222-234.
36.	Pawan P, Dhiman R. Electroencephalogram channel selection based on Pearson correlation coefficient for motor imagery-brain-computer interface[J]. Meas Sens, 2022, 25: 100616.
37.	Thibeault C M, Srinivasa N. Using a hybrid neuron in physiologically inspired models of the basal ganglia[J]. Front Comput Neurosci, 2013, 7: 88.
38.	An M, Le Q. EfficientNet: rethinking model scaling for convolutional neural networks[C]//International Conference on Machine Learning (ICML). Long Beach: PMLR, 2019: 6105-6114.
39.	He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016: 770-778.
40.	Huang G, Liu Z, Van Der Maaten L, et al. Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE, 2017: 4700-4708.
41.	Slenghi F, Sassi R, Rivolta M W. Recovery from coma after cardiac arrest: which time-window counts the most for deep learning predictions?[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.

1. 茶興增, 張月, 張翼飛, 等. 基于兩階卷積神經網絡訓練有限心電數據的心臟驟停早期分類識別算法[J]. 生物醫學工程學雜志, 2024, 41(4): 692-699.
2. Zheng W L, Amorim E, Jing J, et al. Predicting neurological outcome from electroencephalogram dynamics in comatose patients after cardiac arrest with deep learning[J]. IEEE Trans Biomed Eng, 2021, 69(5): 1813-1825.
3. 張唐娟, 張強, 胡悅朋, 等. 構建區域性急診體外心肺復蘇標準流程的實踐及探索[J]. 中華急診醫學雜志, 2025, 34(3): 286-291.
4. Zhou G, Wang Y, Sun Z, et al. Survival outcome among patients with out-of-hospital cardiac arrest who received cardiopulmonary resuscitation in China: a systematic review and meta-analysis[J]. Eur J Med Res, 2023, 28(1): 8.
5. Del Rios M, Bartos J A, Panchal A R, et al. Part 1: executive summary: 2025 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care[J]. Circulation, 2025, 152(16_suppl_2): S284-S312.
6. Kramer P, Miera O, Berger F, et al. Prognostic value of serum biomarkers of cerebral injury in classifying neurological outcome after paediatric resuscitation[J]. Resuscitation, 2018, 122: 113-120.
7. Scholefield B R, Tijssen J, Ganesan S L, et al. Prediction of good neurological outcome after return of circulation following paediatric cardiac arrest: a systematic review and meta-analysis[J]. Resuscitation, 2025, 207: 110483.
8. 吳雪梅, 劉薇薇, 徐莉. 基于BIS監測的鎮靜護理干預在顱腦損傷患者中的應用效果[J]. 現代醫藥衛生, 2024, 40(24): 4230-4233.
9. Young G B. Continuous EEG monitoring in post-cardiac arrest patients: further prognostic insights[J]. Resuscitation, 2016, 109: 114.
10. Khazanova D, Douglas V C, Amorim E. A matter of timing: EEG monitoring for neurological prognostication after cardiac arrest in the era of targeted temperature management[J]. Minerva Anestesiol, 2021, 87(6): 704-713.
11. Aellen F M, Alnes S L, Loosli F, et al. Auditory stimulation and deep learning predict awakening from coma after cardiac arrest[J]. Brain, 2023, 146(2): 778-788.
12. Pham S D T, Keijzer H M, Ruijter B J, et al. Outcome prediction of postanoxic coma: a comparison of automated electroencephalography analysis methods[J]. Neurocrit Care, 2022, 37(Suppl 2): 248-258.
13. Krumm L, Kranz D D, Halimeh M, et al. Toward a universal map of EEG: a semantic, low-dimensional manifold for EEG classification, clustering, and prognostication[J]. Ann Neurol, 2025, 98(2): 357-368.
14. Noirhomme Q, Lehembre R, Lugo Z D R, et al. Automated analysis of background EEG and reactivity during therapeutic hypothermia in comatose patients after cardiac arrest[J]. Clin EEG Neurosci, 2014, 45(1): 6-13.
15. Zheng W L, Amorim E, Jing J, et al. Predicting neurological outcome in comatose patients after cardiac arrest with multiscale deep neural networks[J]. Resuscitation, 2021, 169: 86-94.
16. Zabihi M, Zar A C, Grover P, et al. Hyperensemble learning from multimodal biosignals to robustly predict functional outcome after cardiac arrest[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.
17. Kim D K, Yoon H C, Kim H S, et al. Predicting neurological outcome after cardiac arrest using a pretrained model with electroencephalography augmentation[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.
18. Wang B, Zhang H, Yan M, et al. Predicting neurological recovery after cardiac arrest from electroencephalogram using residual network and random forest[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.
19. Jin J, Miao Y, Daly I, et al. Correlation-based channel selection and regularized feature optimization for MI-based BCI[J]. Neural Netw, 2019, 118: 262-270.
20. Ding X, Zhang X, Ma N, et al. RepVGG: making VGG-style ConvNets great again[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville: IEEE, 2021: 13733-13742.
21. Zhao M, Zhong S, Fu X, et al. Deep residual shrinkage networks for fault diagnosis[J]. IEEE Trans Ind Inf, 2019, 16(7): 4681-4690.
22. Li Y, Guo L, Liu Y, et al. A temporal-spectral-based squeeze-and-excitation feature fusion network for motor imagery EEG decoding[J]. IEEE Trans Neural Syst Rehabil Eng, 2021, 29: 1534-1545.
23. Reyna M A, Amorim E, Sameni R, et al. Predicting neurological recovery from coma after cardiac arrest: the George B. Moody PhysioNet Challenge 2023[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.
24. Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[PP/OL]. arXiv (2014-09-04) [2025-06-16]. https://doi.org/10.48550/arXiv.1409.1556.
25. Donoho D L. De-noising by soft-thresholding[J]. IEEE Trans Inf Theory, 2002, 41(3): 613-627.
26. Isogawa K, Ida T, Shiodera T, et al. Deep shrinkage convolutional neural network for adaptive noise reduction[J]. IEEE Signal Process Lett, 2017, 25(2): 224-228.
27. Li Y, Liu Y, Cui W G, et al. Epileptic seizure detection in EEG signals using a unified temporal-spectral squeeze-and-excitation network[J]. IEEE Trans Neural Syst Rehabil Eng, 2020, 28(4): 782-794.
28. Cong Z, Zhao M, Ling L, et al. A temporal-spectral based single-lead electroencephalogram feature fusion network may provide potential clinical biomarker for cardiac arrest[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.
29. Herman P, Prasad G, McGinnity T M, et al. Comparative analysis of spectral approaches to feature extraction for EEG-based motor imagery classification[J]. IEEE Trans Neural Syst Rehabil Eng, 2008, 16(4): 317-326.
30. Li Y, Wang X D, Luo M L, et al. Epileptic seizure classification of EEGs using time–frequency analysis based multiscale radial basis functions[J]. IEEE J Biomed Health Inform, 2017, 22(2): 386-397.
31. Li Y, Cui W G, Huang H, et al. Epileptic seizure detection in EEG signals using sparse multiscale radial basis function networks and the Fisher vector approach[J]. Knowl Based Syst, 2019, 164: 96-106.
32. Zhang Y, Liu B, Ji X, et al. Classification of EEG signals based on autoregressive model and wavelet packet decomposition[J]. Neural Process Lett, 2017, 45: 365-378.
33. Cox D D, Savoy R L. Functional magnetic resonance imaging (fMRI) "brain reading": detecting and classifying distributed patterns of fMRI activity in human visual cortex[J]. Neuroimage, 2003, 19(2): 261-270.
34. Li Y, Namburi P, Yu Z, et al. Voxel selection in fMRI data analysis based on sparse representation[J]. IEEE Trans Biomed Eng, 2009, 56(10): 2439-2451.
35. Upadhyay R, Manglick A, Reddy D K, et al. Channel optimization and nonlinear feature extraction for electroencephalogram signals classification[J]. Comput Electr Eng, 2015, 45: 222-234.
36. Pawan P, Dhiman R. Electroencephalogram channel selection based on Pearson correlation coefficient for motor imagery-brain-computer interface[J]. Meas Sens, 2022, 25: 100616.
37. Thibeault C M, Srinivasa N. Using a hybrid neuron in physiologically inspired models of the basal ganglia[J]. Front Comput Neurosci, 2013, 7: 88.
38. An M, Le Q. EfficientNet: rethinking model scaling for convolutional neural networks[C]//International Conference on Machine Learning (ICML). Long Beach: PMLR, 2019: 6105-6114.
39. He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016: 770-778.
40. Huang G, Liu Z, Van Der Maaten L, et al. Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE, 2017: 4700-4708.
41. Slenghi F, Sassi R, Rivolta M W. Recovery from coma after cardiac arrest: which time-window counts the most for deep learning predictions?[C]//2023 Computing in Cardiology (CinC). Maastricht: IEEE, 2023, 50: 1-4.

Journal of Biomedical Engineering

A method for predicting neurological outcomes after cardiac arrest based on time-frequency domain feature fusion

Abstract Full text Figures/Tables Video References Cited by

Previous Article

Next Article

Format

Content