基于Logistic回归和局部多项式回归的疾病风险预测

Disease risk Prediction Based on Logistic Regression and Local Polynomial Regression

出血性脑卒中一种危险的神经系统疾病,由脑部血管破裂引起出血,病情急剧进展,病死率高,对患者和社会带来沉重负担。因此,研究出血性脑卒中的诊疗至关重要,可改善患者预后、减少残疾和死亡率,提高医疗系统效率和质量。本文利用Logistic回归、局部多项式回归对血肿扩张和血肿周围水肿两个指标建模,研究出血性脑卒中患者血肿扩张风险、血肿周围水肿发生及演进规律,最终结合临床和影像信息,预测出血性脑卒中患者的临床预后,并据此优化临床决策。

Hemorrhagic stroke is a dangerous neurological disease with bleeding caused by rupture of a blood vessel in the brain, which is rapidly progressive and has a high mortality rate, imposing a heavy burden on patients and society. Therefore, it is crucial to study the diagnosis and treatment of hemorrhagic stroke to improve patient prognosis, reduce disability and mortality, and improve the efficiency and quality of the healthcare system. In this paper, we used Logistic regression and local polynomial regression to model two indicators, hematoma expansion and perihematoma edema, to study the risk of hematoma expansion, the occurrence and evolution of perihematoma edema in hemorrhagic stroke patients, and finally to predict the clinical prognosis of hemorrhagic stroke patients by combining clinical and imaging information, and to optimize clinical decision-making accordingly.