CTAI分析系统与Philips CT后处理工作站在预测急性脑梗死预后中的一致性分析

Consistency Analysis of CTAI Analysis System and Philips CT Post-processing Workstation in Predicting Prognosis of Acute Cerebral Infarction

目的探讨CT脑灌注成像(CTP)智能(AI)分析系统与Philips CT后处理工作站在预测急性脑梗死(ACI)预后中的一致性。方法选取我院2021年6月至2023年6月ACI患者100例,根据90 d预后情况分为预后良好组(77例)与预后不良组(23例)。入院时均行CTP检查,比较AI分析系统与Philips CT后处理工作站测量的患侧最大层面与对应健侧血流参数[脑血流量(CBF)、脑血容量(CBV)、平均通过时间(MTT)、相对脑血流量(rCBF)、相对脑血容量(rCBV)、相对平均通过时间(rMTT)、缺血区最大层面CBF、CBV面积]预测ACI患者预后的一致性。结果AI分析系统、Philips CT后处理工作站测量的预后不良组患侧最大层面CBF、CBV、rCBF、rCBV低于预后良好组,患侧最大层面MTT、rMTT、缺血区最大层面CBF、CBV面积高于预后良好组(P<0.05);AI分析系统、Philips CT后处理工作站测量的两组患侧最大层面与对应健侧CBF、CBV、MTT、rCBF、rCBV、rMTT、缺血区最大层面CBF、CBV面积差异无统计学意义(P>0.05);患侧最大层面CBF、CBV、MTT、rCBF、rCBV、rMTT、缺血区最大层面CBF、CBV面积均是ACI预后的影响因素(P<0.05);AI分析系统测量的患侧最大层面CBF、CBV、MTT、rCBF、rCBV、rMTT、缺血区最大层面CBF、CBV面积预测ACI预后AUC分别为0.742、0.776、0.842、0.809、0.782、0.755、0.742、0.787;Philips CT后处理工作站测量的患侧最大层面CBF、CBV、MTT、rCBF、rCBV、rMTT、缺血区最大层面CBF、CBV面积预测ACI预后的AUC分别为0.742、0.722、0.816、0.803、0.772、0.761、0.750、0.781;两种方式测量出的CTP参数预测ACI预后的AUC差异均无统计学意义(P>0.05),一致性Kappa值均>0.7。结论CTP AI分析系统与Philips CT后处理工作站测量的血流参数在预测ACI预后中一致性良好,可将AI分析系统作为优选的图像后处理方式。

Objective To investigate the consistency of intelligent(AI)analysis system of CT cerebral perfusion imaging(CTP)and Philips CT post-processing workstation in predicting the prognosis of acute cerebral infarction(ACI).Methods A total of 100 ACI patients from our hospital from June 2021 to June 2023 were selected and divided into a good prognosis group(77 cases)and a poor prognosis group(23 cases)based on the 90 day prognosis.At admission,CTP examination was performed to compare the consistency of predicting the prognosis of ACI patients by comparing the maximum cross-sectional blood flow parameters(CBF,CBV,MTT,rCBF,rCBV,rCVT,rCVT,and CBV area)measured by AI analysis system and Philips CT post-processing workstation on the affected side and corresponding healthy side.Results The maximum layer CBF,CBV,rCBF and rCBV of the affected side in the poor prognosis group were lower than those in the good prognosis group,and the maximum layer MTT,rMTT and the maximum layer CBF and CBV of the ischemic area were higher than those in the good prognosis group(P<0.05).There was no significant difference in CBF,CBV,MTT,rCBF,rCBV,rMTT,and the area of CBF and CBV at the maximum layer of ischemic zone between the two groups measured by AI analysis system and Philips CT post-processing workstation(P>0.05).The maximum layer CBF,CBV,MTT,rCBF,rCBV,rMTT,and the maximum layer CBF and CBV area of ischemic area were all influencing factors for the prognosis of ACI(P<0.05).The AUC of the maximum layer CBF,CBV,MTT,rCBF,rCBV,rMTT,and the area of the maximum layer CBF and CBV measured by AI analysis system to predict the prognosis of ACI were 0.742,0.776,0.842,0.809,0.782,0.755,0.742,and 0.787,respectively.Philips The AUC of CBF,CBV,MTT,rCBF,rCBV,rMTT,and the area of CBF and CBV at the maximum layer of the affected side to predict the prognosis of ACI measured by CT post-processing workstation were 0.742,0.722,0.816,0.803,0.772,0.761,0.750 and 0.781,respectively 1;The CTP parameters measured by the two methods had no statistical significance in predicting ACI prognosis AUC(P>0.05),and the consistency Kappa value was greater than 0.7.Conclusion The blood flow parameters measured by CTP AI analysis system and Philips CT post-processing workstation were in good agreement in predicting the prognosis of ACI,and AI analysis system could be used as the preferred post-processing method.