AUTHOR=Hu Ping , Gao Yanjuan , Zhang Yiqian , Sun Kui 

TITLE=Ultrasound image-based deep learning to differentiate tubal-ovarian abscess from ovarian endometriosis cyst

JOURNAL=Frontiers in Physiology

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/physiology/articles/10.3389/fphys.2023.1101810

DOI=10.3389/fphys.2023.1101810

ISSN=1664-042X

ABSTRACT=Abstract
Objectives: We developed ultrasound (US) images-based convolutional neural networks (CNNs) to distinguish between tubal-ovarian abscess (TOA) and ovarian endometriosis cyst (OEC).
Methods: A total of 202 patients who underwent US scanning and confirmed TOA or OEC by pathology were enrolled in retrospective research, in which 171 patients (from Jan 2014 to Sep 2021) were considered primary cohort (training, validation and internal test sets) and 31 patients (from Sep 2021 to Dec 2021) were considered independent test cohort. There were 68 TOAs and 89 OECs, 4 TOAs and 10 OECs, 10 TOAs and 21 OECs patients belonging to training and validation sets, internal, and independent test sets, respectively. For the model to gain better generalization, we applied the geometric image and color transformations to augment the dataset, including center crop, random rotation, random horizontal flip, etc. Three CNNs included ResNet-152, DenseNet-161 and EfficientNet-b7 were applied to differentiate TOA from OEC and compared performance with the three US physicians and clinical indicator of carbohydrate antigen 125 (CA125) on the independent test set. The area under the receiver operating characteristic curves (AUROCs), accuracy, sensitivity, and specificity were used to evaluate the performance.
Results: In the CNNs, The performance of ResNet-152 was the highest, with AUROCs of 0.986 (0.954-1). The AUROCs of the three physicians were 0.781 (0.620-0.942), 0.738 (0.629-848), and 0.683 (0.501-0.865), respectively. The clinical indicator CA125 achieved only 0.564 (0.315-0.813).
Conclusion: We demonstrated the CNNs model based on the US image could discriminate TOA and OEC better than US physicians and CA125. This method can provide a valuable predictive reference for physicians to screen TOA and OEC in time.