Magnetic susceptibility imaging combined with transcranial sonography is good in predicting Parkinson's disease
Objective To explore the diagnostic value of susceptibility-weighted imaging(SWI)combining transcranial sonography(TCS)in Parkinson's disease(PD).Methods Forty-two PD patients treated at the First People's Hospital of Foshan from January 2021 to December 2023 were selected as the PD group,while 43 healthy volunteers served as the control group.All subjects underwent SWI and TCS examinations to observe the disappearance of the"swallow tail sign"in the midbrain substantia nigra.Differences in the hyperechoic area of the substantia nigra(SN),midbrain area,and the S/M values were analyzed between the groups.Statistically significant indicators were subjected to multivariate logistic regression analysis to identify independent predictors of PD and construct a predictive model,which was visualized using a nomogram.The model's discriminative ability,calibration,and clinical utility were evaluated using ROC curves,calibration curves,and decision curve analysis,respectively.Results There were significant differences in the hyperechoic area of the SN(P＜0.001)and the S/M ratio(P＜0.001)between the PD group and the control group.The AUC for SWI and TCS separately in distinguishing PD from control group were 0.80(95%CI:0.716-0.886)and 0.857(95%CI:0.774-0.940),respectively.Multivariate Logistic regression analysis showed that the hyperechoic area of SN and the swallow tail sign as independent risk factors for predicting PD.The predictive model constructed using these factors had an AUC value of 0.94(95%CI:0.896-0.984).The calibration curve demonstrated good agreement between predicted and actual probabilities,and decision curve analysis showed that the predictive model provided a higher net benefit than individual indicators.Conclusion The predictive model combining SWI and TCS quantitative indicators demonstrates good predictive capability for PD risk,surpassing any single model and showing promising clinical utility.
国家科技期刊平台
NSTL
万方数据
