Multiple model parameters of intravoxel incoherent motion in differential diagnosis of cervical carcinoma and pre-judgement of pathological types / 中国医学影像技术

ABSTRACT

Objective To investigate the feasibility of multiple model parameters of intravoxel incoherent motion (IVIM) in differential diagnosis of cervical carcinoma and its pathological types.Methods IVIM images of 33 patients with cervical cancer (cervical cancer group) and 30 patients with normal cervix (normal group) were retrospectively analyzed.The monoexponential model parameters (ADC-stand),biexponential model parameters (ADC-slow,ADC-fast,f) and stretched-exponential model values (distributed diffusion coefficient [DDC],α) were measured,respectively.Then the parameters were compared among the two groups,as well as among different pathological types and different pathological grades of cervical cancer.The thresholds and diagnostic efficiency of the parameter values were evaluated with ROC curve.Results ADC-stand,ADC-slow,f,DDC and α values of cervical cancer group were lower than those of normal group (all P＜0.05).The area under the ROC curve of ADC-stand,ADC-slow,DDC,f and a values in diagnosis of cervical cancer was 0.93,0.94,0.96,0.94 and 0.93 (all P＜0.05),and the diagnostic thresholds were 1.03 × 10-3 mm2/s,0.99 × 10-3mm2/s,30.50％,1.21 × 10-3 mm2/s and 0.699,respectively.ADC-stand,ADC-slow and DDC values in cervical squamous cell carcinoma patients were lower than those of cervical adenocarcinoma patients (all P＜0.05).The area under the ROC curve of ADC-stand,ADC-slow and DDC values in differential diagnosis of cervical squamous cell carcinoma and cervical adenocarcinoma was 0.98,0.91 and 0.98 (all P＜0.05),the diagnostic thresholds were 1.11 × 10-3 mm2/s,0.81×10-3 mm2/s and 0.98 × 10-3 mm2/s,respectively.There was no significant difference of the parameters among high,moderate,low differentiation of cervical squamous cell carcinoma.Conclusion Multiple model parameters of IVIM can quantitatively reflect the histological features of cervical cancer.