Objective To explore the value of the corrected-size ratio(c-SR)value of intracranial volume computed tomography angiography(CTA)in predicting the risk of intracranial aneurysm rupture.Methods A total of 81 patients with aneurysms who had follow-up records were selected.Among them,39 patients with unruptured aneurysms and underwent regular follow-ups,while 9 patients with unruptured aneurysms opted for surgical intervention.Additionally,surgery was performed on 33 patients with ruptured aneu-rysms.Three-dimensional reconstruction of CTA was performed to obtain the morphological parameters of aneurysms.The initial size ratio(SR)value of aneurysm and the follow-up SR value or postoperative c-SR value were obtained.The changes in SR values of unruptured aneurysms were analyzed,the preoperative SR values and postoperative c-SR values of aneurysms were analyzed.Finally,the correla-tion between the intial SR value of unruptured aneurysms and the c-SR value of ruptured aneurysms was compared.Results No sig-nificant difference was observed between the initial SR value of unruptured aneurysms and the follow-up SR value(P>0.05).Simi-larly,no significant difference was noted between the preoperative SR value of unruptured aneurysms and the postoperative c-SR value(P>0.05).The preoperative SR value of ruptured aneurysms differed significantly from the postoperative c-SR value(P<0.05).There was a significant difference between the initial SR value of unruptured aneurysms and the postoperative c-SR value of ruptured aneurysms(P<0.05).The receiver operating characteristic(ROC)curve analysis was performed on the initial SR value of unrup-tured aneurysms and the postoperative c-SR value of ruptured aneurysms.The area under the curve(AUC)was 0.860 and the best cut-off value was 1.045.Conclusion Unruptured aneurysms remain stable for an extended period of time,exhibiting no significant change in morphological parameters.It can be concluded that surgical intervention does not affect the SR value of aneurysms.In the case of subarachnoid hemorrhage caused by ruptured aneurysms,the parent artery become thinner,then the preoperative SR value of ruptured aneurysms may be exaggerated,which results in the distortion of the preoperative SR value of ruptured aneurysms.However,the postoperative c-SR value is the true SR value before the rup-ture of aneurysms.

Objective To explore the value of three-dimensional pseudo-continuous arterial spin labeling(3D pCASL)and diffu-sion weighted imaging(DWI)in the differential diagnosis of stage T1 nasopharyngeal carcinoma(NPCT1)and lymphoid hyperplasia(LH).Methods A total of 21 patients with pathological diagnosis and clinical stage of NPCT1(NPCT1 group)and 50 patients with pathological diagnosis of LH(LH group)were selected.All patients underwent nasopharyngeal 3D pCASL and DWI scans before treatment.The blood flow(BF)values of all lesions[minimum BF(BFmin),mean BF(BFmean),maximum BF(BFmax)],and the rel-ative blood flow(rBF)values of the ratio of lesions to lateral pterygoid muscle at the same plane[minimum rBF(rBFmin),mean rBF(rBFmean),maximum rBF(rBFmax)],the apparent diffusion coefficient(ADC)values[minimum ADC(ADCmin),mean ADC(ADCmean),maximum ADC(ADCmax)]of all lesions were measured.The differences in parameters between NPCT1 group and LH group were analyzed,and the diagnostic efficiency of each parameter was analyzed via receiver operating characteristic(ROC)curve.Results The values of BFmean,BFmax,rBFmin,rBFmean and rBFmax of NPCT1 group were higher than those of LH group,with statisti-cally significant difference(P<0.05).However,there were no significant difference in the values of ADCmin,ADCmean and ADCmax between the two groups(P>0.05).The area under the curve(AUC)of BFmean,BFmax,rBFmin,rBFmean and rBFmax values for differen-tial diagnosis of NPCT1 and LH were 0.677,0.804,0.748,0.746 and 0.858,respectively.Conclusion 3D pCASL technique can reflect non-invasively the difference of blood perfusion between NPCT1 and LH,and can be used as an effective method to distin-guish NPCT1 from LH,with the better diagnostic efficiency of BFmax and rBFmax.However,DWI is difficult to distinguish the difference of water molecule diffusion between NPCT1 and LH,which has limited value in differential diagnosis.

Objective To investigate the CT and MRI findings of rhabdomyosarcoma(RMS).Methods A retrospective analysis was conducted on CT and/or MRI data of 19 cases of pathologically confirmed RMS.The growth location,morphology,density/sig-nal intensity,enhancement pattern,and infiltration scope of the lesions were analyzed.Results Among the 19 patients,12 lesions were located in the head and neck regions(sinuses,nasal cavity,masseter muscle,infratemporal fossa,nasopharynx,and gum),2 in the lower extremities,1 in the left heart,1 in the mediastinum,1 in the prostate,1 in the testicle,and 1 in the pelvis.On CT scans,the lesions appeared as homogeneously iso-or hypodensity compared to the muscle,with homogeneous or inhomogeneous enhancement on contrast-enhanced scans.On MRI,the lesions exhibited iso-or hypointense signals on T1WI and inhomogeneous hyperintense signals on T2WI,with inhomogeneous enhancement on contrast-enhanced scans.All lesions had irregular morphologies but clear boundaries,and no calcification or hemorrhage was observed within the lesions.Lesions in the sinuses often invaded adjacent sinus cavities or the nasal cavity and orbit,and some lesions could involve the intracranial cavity through the cranial base foramina.Bone destruction could occurred.Regional lymph node enlargement in the drainage area was observed in some cases.One case exhibited multiple osseous metastases.Conclusion RMS can occur in various parts of the body,with the head and neck being the most common site.Most cases lack specificity,but CT and MRI is helpful to understand the location of the disease,its invasion of surrounding tissues,and the presence of metastasis,thereby assisting in treatment and prognosis evaluation.

Objective To explore the predictive value of quantitative parameters of dual-layer detector spectral CT(DLCT)for Ki-67 expression in solid lung adenocarcinoma.Methods The data of 103 patients were retrospectively collected,and the patients were divided into Ki-67 high expression and Ki-67 low expression groups according to Ki-67 proliferation index.The quantitative parame-ters of DLCT were measured and calculated,and the differences in these parameters between the two groups were compared.The parameters with statistically significant differences were assessed for correlation with Ki-67 expression.The efficacy of DLCT param-eters and combined parameters in predicting Ki-67 expression in solid lung adenocarcinoma were evaluated by receiver operating character-istic(ROC)curve and compared by DeLong test.Results Long diameter,short diameter and smoking history were positively corre-lated with the Ki-67 expression in solid lung adenocarcinoma(r>0,P<0.05).Gender and quantitative parameters of DLCT were nega-tively correlated with the Ki-67 expression in solid lung adenocarcinoma(r<0,P<0.05).The combined parameters of convention and spectral CT had the highest prediction efficiency.Conclusion The quantitative parameters of DLCT can be used to evaluate the Ki-67 expres-sion in solid lung adenocarcinoma.

Objective To investigate the value of constructing a risk prediction model of brain metastasis in lung cancer based on clinical-CT imaging.Methods The clinical and CT imaging data of 208 patients with lung cancer confirmed by surgical pathology or puncture biopsy were analyzed retrospectively,including 98 patients in the metastasis group and 110 patients in the non-metastasis group.Univariable and binary logistic regression analyses were performed between the two groups,and the clinical,CT imaging,and clinical-CT imaging models were constructed according to the selected independent risk factors.Prediction model performance was eval-uated with receiver operating characteristic(ROC)curve,calibration curve and decision curve analysis(DCA).Results Multivariate analysis showed that T stage,pathological type,radiotherapy and chemotherapy,surgery,long diameter(LD),short diameter(SD),minimum CT value(CTmin)were the independent risk factors for predicting brain metastasis in lung cancer(P<0.05).The area under the curve(AUC)of clinical,CT imaging and clinical-CT imaging models were 0.925,0.764,0.941,respectively.DeLong test analysis showed that the AUC of clinical-CT imaging model,clinical model and CT imaging model was statistical difference(Z=2.093,5.777,all P<0.05).The calibration curve suggested a good fit of the clinical-CT imaging model.The DCA suggested that the clinical-CT imaging model demonstrates good clinical benefits.Conclusion The clinical-CT imaging model can effectively predict the occurrence of brain metastasis in lung cancer,which is helpful to guide the development of accurate diagnosis and treatment plan.

Objective To investigate the correlation of the cumulative exposure of body mass index(BMI)with epicardial adipose tissue(EAT)volume.Methods Based on the Kailuan study cohort,subjects were divided into Q1,Q2 and Q3 according to the cumula-tive BMI exposure level.The generalized linear regression model was used to analyze the correlation between different cumulative BMI exposure levels and EAT volume.According to the median of EAT volume(115.83 cm3),EAT volume was defined as increased EAT volume when EAT volume was higher than or equal to the median EAT volume.Logistic regression model was used to analyze the correlation between different cumulative BMI exposure levels and increased EAT volume.Results With the increased of cumu-lative BMI exposure level.The EAT volume increased gradually and the risk of increased EAT volume was significantly increased.Conclusion The level of cumulative BMI exposure is significantly correlated with EAT volume.High levels of cumulative BMI exposure lead to an increased risk of increased EAT volume.

Objective To investigate the diagnostic value of dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI)combined with Doppler ultrasound and elastography in breast lesions.Methods A total of 91 patients with breast lesions confirmed by pathology were retrospectively selected and underwent DCE-MRI,Doppler ultrasound and elastography examinations.The imaging features of benign and malignant breast lesions were analyzed.The diagnostic value of DCE-MRI,Doppler ultrasound,and elastogra-phy in benign and malignant breast lesions were analyzed.Results Among the 91 cases,there were 62 cases with benign lesions(benign group)and 29 cases with malignant lesions(malignant group).There were statistically significant differences in DCE-MRI parameters,Doppler ultrasound parameters and elastography parameters between the two groups(P<0.05).Binary logistic regres-sion analysis found that enhancement characteristics,morphology,time-signal intensity curve(TIC),blood flow signals on Doppler ultrasound,elastography score,and strain rate(SR)were the influencing factors of benign and malignant breast lesions(P<0.05).The differential diagnostic efficacy of combination of the three examinations was highest.The sensitivity,specificity and accuracy were 88.71%,82.76%and 86.81%,respectively.Conclusion DCE-MRI,Doppler ultrasound,and elastography are effective for differentiating benign and malignant breast lesions.Combination of them can improve the diagnostic performance.

Objective To investigate the value of quantitative parameters of diffusion weighted imaging(DWI)based on mono-expo-nential model(MEM),diffusion kurtosis imaging(DKI)model,and stretched-exponential model(SEM)in predicting the efficacy of neoadjuvant therapy in locally advanced gastric cancer(LAGC).Methods Forty LAGC patients who underwent MRI examinations before neoadjuvant therapy and before radical surgery were prospectively enrolled.A radiologist delineated lesions on DWI images and acquired quantitative parameters before and after treatment,including lesion volume,apparent diffusion coefficient(ADC)of MEM,mean diffusivity(MD)and mean kurtosis(MK)of DKI model,distribution diffusion coefficient(DDC),and α of SEM.According to pathological tumor regression grade(TRG),the patients were stratified into good response group(TRG 0-1)and poor response group(TRG 2-3).The pre-treatment parameters and Δ of pre-and post-treatment parameters were compared between the two groups with Mann-Whitney U test;multivariate analysis was performed with binary logistic regression.Multiple DWI models and the combined model were established,and the prediction efficiency of each model was calculated.Results There was no significant differ-ence in each parameter before neoadjuvant therapy between the two groups(P>0.05).The delta of volume,ΔADC,ΔMD,and ΔDDC pre-and post-treatment were all statistically different between the two groups(P<0.05).The area under the curve(AUC)of ΔADC,ΔMD,and ΔDDC in predicting good response for LAGC were 0.900,0.806,and 0.762,respectively.The AUC of the combined model was 0.946.Conclusion Quantitative parameters of MEM,DKI model,and SEM can help predict the efficacy of neoadju-vant-treated LAGC patients.

Objective To explore the value of radiomics model and deep learning model based on unenhanced MRI in predicting microvascular invasion(MVI)of hepatocellular carcinoma(HCC)preoperatively.Methods A total of 189 patients with postopera-tive pathologically confirmed HCC from two centers were retrospectively selected,of which 119 cases from Zhujiang Hospital of Southern Medical University were used as the training set[60 cases with negative MVI,59 cases with positive MVI],and 70 cases from Shenzhen People's Hospital were used as the external test set[38 cases with negative MVI and 32 cases with positive MVI].Clinical indicators were analyzed by univariate and multivariate logistic regression analysis and the independent predictors of positive MVI were screened.Deep transfer learning(DTL)and traditional radiomics methods were used to construct radiomics model and deep learning model based on unenhanced MRI.The predictive performances of each model were compared using receiver operating charac-teristic(ROC)curves and area under the curve(AUC).DeLong test was employed to compare statistical differences in performance of the models.Results Alkaline phosphatase(ALP)and prothrombin time(PT)were independent predictors of positive MVI(P<0.05).The deep learning model based on T2WI had the best predictive efficacy,with AUC of 0.779[95%confidence interval(CI)0.696-0.863]and 0.741(95%CI 0.620-0.861)in the training set and external test set,respectively,and there were statistically significant differences compared with the radiomics model and the clinical model based on T1WI(P<0.05).Conclusion Deep learning model based on T2WI has a certain application value in preoperative noninvasive prediction of MVI status in HCC patients.

Objective To explore the clinical application value of gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid(Gd-EOB-DTPA)enhanced MRI T1 mapping in the evaluation of liver function.Methods Sixty-four patients who underwent enhanced MRI T1 mapping with Gd-EOB-DTPA and completed the laboratory examination of liver function within one week were prospectively enrolled.All patients were divided into normal control group(NCG),cirrhosis Child-Pugh A(CCA)group,cirrhosis Child-Pugh B(CCB)group,and cirrhosis Child-Pugh C(CCC)group.CCB+CCC groups were defined as a moderate and severe abnormal liver function group.The T1 mapping images of pre-enhanced,post-enhanced 10 min and 20 min were collected,and the T1 mapping val-ues of liver and spleen were measured.The ΔT1 and hepatocyte enhancement fraction(HEF)were calculated.The differences of parameters in different liver function groups were analyzed and compared,and the diagnostic efficacy of each index in distinguishing different liver function groups was evaluated.Results There were significant differences in T1plain scan,T110 min,T120 min,ΔT110 min,ΔT120 min,HEF10 min and HEF20 min among the three groups(P<0.05).The difference of T1plain scan between NCG and CCA groups,and between NCG and CCB+CCC groups was statistically significant(P<0.05).The area under the curve(AUC)of differentiating normal liver function group from abnormal liver function group was 0.761.There were significant differences in T110 min,T120 min,ΔT110 min,ΔT120 min,HEF10 min and HEF20 min between CCA and CCB+CCC groups.The AUC of differentiating the two groups was 0.757,0.820,0.735,0.820,0.790 and 0.853,respectively,and HEF20 min had the highest diagnostic efficacy.Conclusion Gd-EOB-DTPA enhanced MRI T1 mapping can be used as an effective method to evaluate liver function.