RESUMO
Objective To analyze the characteristics of multi-b-value diffusion weighted imaging (DWI) of the normal cervical, cervical cancer and cervical intraepithelial neoplasia(CIN Ⅲ),and to assess the value of DWI in diagnosis and pathological staging for cervical cancer.Methods There were 54 patients with pathologically confirmed cervical cancer,including 36 squamous cell carcinoma,15 adenomatous carcinoma, and 3 neuroendocrine carcinoma.And other 9 patients with CIN Ⅲ were involved in the study.All patients underwent conventional MR scan with multi-b-value DWI sequence.The apparent diffusion coefficient (ADC) values of multi-b-value DWI were compared between normal cervical tissue and cervical cancer with its different pathology subtypes.Results The three normal structure of uterine cervix, including cervical endometrial, junctional zone and myometrium, can be clearly distinguished by different b-value DWI, with a statistical difference of ADC value (P<0.05).Further analysis showed that the average ADC values between normal cervix and cervical cancer had a statistical difference (P<0.05).The similar situation was presented among squamous cell carcinoma, adenomatous carcinoma, neuroendocrine carcinoma and CIN Ⅲ(P<0.05).Conclusion The ADC values of multi-b-value DWI could be helpful to diagnose cervical cancer and CIN Ⅲ,and it is correlated with the pathological type of cervical cancer.The application of multi-b-value DWI to cervical cancer and CIN Ⅲ could increase the accuracy of diagnosis.
RESUMO
Objective To evaluate the diagnostic value of color Doppler ultrasound (CDUS ) in thyroid microcarcinom (TMC). Methods With surgical pathologic examination results as the gold standard, 114 thyroid lumps were divided in-to observation group (including 51 pieces of TMC lump)and control group (including 63 pieces of benign lump). The sign of CDUS were analyzed retrospectively in the two groups. Logistic regression analysis was used to screen the CDUS characteristic parameters which associated with TMC and established TMC diagnosis prediction model. Results The percentage of anteroposterior and transverse diameter ratio(A/T)>1, unclear edge, hypoechoic nodules, absent halo sign, micro calcifications, resistance index(RI)>0.70 of the observation group were higher than those of the control group(P<0.01). Blood flow signal distribution in the 2 groups were different significantly (P<0.01). Multi-factor stepwise Logistic regression analysis results show that the absent halo sign, unclear edge, RI>0.70, hypoechoic nodules and blood flow signal ≤2 were associated with TMC (P<0.05). The above five indexes were used to establish regression model to pre-dict TMC and the sensitivity, specific degrees and accuracy were 93.75%, 90.90%, 92.11%. Conclusion The Logistic regression model with echo, edge, RI and blood flow signal in CDUS can be helpful for early accurate diagnosis of TMC.