Search

Application of Deep Learning in Risk Prediction of Cerebral Infarction

Author(s)

Wenjie Jia, Guohong Wang*

Affiliation(s)

Department of Neurosurgery, The First People's Hospital of Jiayuguan City, Gansu, China *Corresponding Author

Abstract

Cerebral infarction is a global high incidence of cerebrovascular disease, with high disability rate, high mortality, high recurrence rate characteristics, serious threat to human health and quality of life. Traditional methods for predicting the risk of cerebral infarction are mostly based on static clinical data and classical statistical models, which are difficult to capture the dynamic temporal changes of patients' health status, and cannot fully explore the potential associations in long-term follow-up data, and the prediction accuracy is obviously limited. Deep learning technology shows significant advantages in the field of time series data processing, among which long-term memory recurrent neural network (LSTM-RNN) effectively solves the gradient disappearance problem of traditional recurrent neural network through gating structure, and can realize accurate extraction of long-term time series features and effective capture of long-term dependence relationship. In this paper, a cerebral infarction risk prediction model based on LSTM-RNN is constructed, standardized collection and systematic pretreatment of time series clinical data are completed, the overall framework of the model and the whole process training scheme are designed, and the prediction performance and generalization ability of the model are verified by multi-dimensional experiments. The experimental results show that the model shows excellent accuracy and robustness in the task of predicting cerebral infarction risk, and each core evaluation index is significantly better than that of traditional statistical model and other recurrent neural network models. It can provide reliable technical support for early screening, risk stratification and accurate prevention of cerebral infarction, and has important clinical application value and social significance.

Keywords

Cerebral Infarction; Risk Prediction; Deep Learning; Long-Term Memory; Recurrent Neural Network; Time Series Clinical Data

References

[1]Li X, Wang Y, Xu J. Development of a machine learning-based risk prediction model for cerebral infarction and comparison with nomogram model[J]. Journal of Affective Disorders, 2022, 314: 341-348. [2]Abdulazim A, Heilig M, Rinkel G, et al. Diagnosis of delayed cerebral ischemia in patients with aneurysmal subarachnoid hemorrhage and triggers for intervention[J]. Neurocritical care, 2023, 39(2): 311-319. [3]Dewi T, Mardiyati E N, Risma P, et al. Hybrid Machine learning models for PV output prediction: Harnessing Random Forest and LSTM-RNN for sustainable energy management in aquaponic system[J]. Energy Conversion and Management, 2025, 330: 119663. [4]Zeng Z, Luo R, Xu W, et al. Prediction of first-onset cerebral infarction risk in patients with acute myocardial infarction: a retrospective cohort Study[J]. International Journal of General Medicine, 2025: 3501-3513. [5]Arafa A, Kokubo Y, Sheerah H A, et al. Developing a stroke risk prediction model using cardiovascular risk factors: the Suita Study[J]. Cerebrovascular Diseases, 2022, 51(3): 323-330. [6]Muehlen I, Sprügel M, Hoelter P, et al. Comparison of two automated computed tomography perfusion applications to predict the final infarct volume after thrombolysis in cerebral infarction 3 recanalization[J]. Stroke, 2022, 53(5): 1657-1664