Objective:

Evaluating the clinical value of IOTA Logistic regression model LR2 in predicting benign and malignant ovarian tumors by ultrasound.

Methods:

This study selected the patients who had been hospitalized surgical treatment in the Third Affiliated Hospital of Zhengzhou University due to the adnexal tumors

all the patients underwent ultrasound

observed and summarized the image

then used IOTA post-processing software calculating the

risk index

follow-up postoperative pathological results

analyzed the sensitivity

specificity

positive predictive value

negative predictive value

positive likelihood ratio

negative likelihood ratio and the area under curve (AUC) of the receiver operating characteristic (ROC).

Results:

A total of 215 patients were calculated

including 126 benign tumors (58.6%) and 89 malignant tumors(41.4%)

the sensitivity was 95.5% (95%CI: 90.4%-98.3%)

the specificity was 76.2% (95%CI: 62.7%-87.7%)

the AUC was 0.89 (SE=0.024

95%CI: 0.87-0.91).

Conclusion:

IOTA Logistic regression model LR2 has highly diagnosis value in evaluating the ovarian tumors

which can be used as an auxiliary diagnosis method for non-gynecological ultrasound or non-expert operators.