Comparison of diagnostic efficiency between IOTA LR2 model and doctors' experiences

Area of Science:

• Gynecologic Oncology
• Medical Imaging
• Diagnostic Accuracy

Background:

• Adnexal masses require accurate diagnosis to differentiate benign from malignant conditions.
• Subjective assessment by clinicians is a common diagnostic method.
• The International Ovarian Tumor Analysis (IOTA) group developed the logistic regression (IOTA LR2) model for risk prediction.

Purpose of the Study:

• To compare the diagnostic efficiency of the IOTA LR2 model against subjective assessment by doctors.
• To evaluate the model's utility for differentiating benign and malignant adnexal masses.

Main Methods:

• Retrospective analysis of 616 adnexal mass ultrasonographic images.
• Comparison between senior and junior doctors' groups using both IOTA LR2 model and subjective assessment.
• Postoperative pathological diagnosis served as the gold standard.

Main Results:

• The IOTA LR2 model demonstrated higher diagnostic accuracy (AUC 0.90 and 0.88) than subjective assessment (AUC 0.86 and 0.79) for senior and junior doctors, respectively.
• Sensitivity and specificity for malignant adnexal masses varied between methods and experience levels.
• No significant difference in sensitivity was found between junior doctors using IOTA LR2 and senior doctors using subjective assessment.

Conclusions:

• The IOTA LR2 model's diagnostic efficiency is on par with experienced senior clinicians.
• The IOTA LR2 model can improve diagnostic accuracy for junior doctors, potentially reducing missed malignant adnexal mass diagnoses.