CT平扫影像特征及临床危险因素预测脑出血血肿扩大的评分系统分析

Analysis of Scoring System for Predicting Hematoma Enlargement in Cerebral Hemorrhage with CT Plain Scan Imaging Features and Clinical Risk Factors

目的分析CT平扫影像特征及临床危险因素预测脑出血血肿扩大的评分系统。方法选择2021年5月—2022年12月本院收治的75例脑出血患者为研究样本。结合患者复查血肿是否扩大,分成未扩大组、扩大组。评价CT平扫影像特征及临床危险因素预测脑出血血肿扩大的评分系统情况。结果所有患者中,出血位点在基底节区域共计43例,脑叶13例、丘脑16例、小脑3例。扩大组和未扩大组的高血压例数、凝血功能障碍例数、首次检查时间、出血量、黑洞征、混合征、黑洞征和混合征例数之间存在统计学意义(P<0.05)。黑洞征和混合征预测脑出血血肿扩大的约登指数、阴性预测值、阳性预测值、特异度、敏感度无统计学意义(P>0.05)。多因素分析结果显示:共有混合征、黑洞征、CT第一次检查时间两个变量被保留到回归方程之中,代表CT首次检查时间、黑洞征和混合征是影响患者血肿扩大独立高危因素。实验将扩大组视为阳性样本,将未扩大组视为阴性样本。创建出CT首次检查时间、黑洞征和混合征ROC诊断分析具体模型。AUC的详细分布为0.626、0.965。结论针对病患开展CT平扫能积极预测大脑血肿扩大情况,其中缩短CT第一次检查时间、混合征以及黑洞征有着比较高的预测价值。

Objective To analyze scoring system for predicting expansion of cerebral hemorrhage and hematoma based on CT plain scan imaging features and clinical risk factors.Methods The paper chose 75 patients with cerebral hemorrhage admitted to our hospital from May 2021 to December 2022 as study samples,and divided them into unexpanded group and enlarged group based on hematoma being expanded or not during reexamination.Scoring system for predicting expansion of cerebral hemorrhage and hematoma based on CT plain scan imaging features and clinical risk factors was evaluated.Results Among all patients,there were a total of 43 cases with bleeding sites in basal ganglia area,13 cases in cerebral lobes,16 cases in thalamus and 3 cases in cerebellum.There was statistical significance between number of cases of hypertension,cases of coagulation dysfunction,time of first examination,bleeding volume,black hole sign,mixed sign,black hole sign and mixed sign in expanded and non expanded groups(P<0.05).There was no statistical significance for Youden's J statistic,negative predictive value,positive predictive value,specificity and sensitivity of black hole sign and mixed sign(P>0.05).Multivariate analysis results showed,mixed signs,black hole sign and CT first examination time,two variables were retained in regression equation,representing that CT first examination time,black hole sign and mixed sign was independent high-risk factors of hematoma enlargement of patients.the experiment treated expanded group as positive samples and non expanded group as negative samples,and created specific models for the first CT examination time,black hole sign and mixed sign ROC diagnostic analysis.Detailed distribution of AUC was 0.626 and 0.965.Conclusion CT plain scan can actively predict expansion of cerebral hematoma of patients,shorten the first CT examination time,mixed sign and black hole sign,which has high predictive value.