Dataset Information

Application of Deep Convolutional Neural Networks for Discriminating Benign, Borderline, and Malignant Serous Ovarian Tumors From Ultrasound Images.

ABSTRACT:

Objective

This study aimed to evaluate the performance of the deep convolutional neural network (DCNN) to discriminate between benign, borderline, and malignant serous ovarian tumors (SOTs) on ultrasound(US) images.

Material and methods

This retrospective study included 279 pathology-confirmed SOTs US images from 265 patients from March 2013 to December 2016. Two- and three-class classification task based on US images were proposed to classify benign, borderline, and malignant SOTs using a DCNN. The 2-class classification task was divided into two subtasks: benign vs. borderline & malignant (task A), borderline vs. malignant (task B). Five DCNN architectures, namely VGG16, GoogLeNet, ResNet34, MobileNet, and DenseNet, were trained and model performance before and after transfer learning was tested. Model performance was analyzed using accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC).

Results

The best overall performance was for the ResNet34 model, which also achieved the better performance after transfer learning. When classifying benign and non-benign tumors, the AUC was 0.96, the sensitivity was 0.91, and the specificity was 0.91. When predicting malignancy and borderline tumors, the AUC was 0.91, the sensitivity was 0.98, and the specificity was 0.74. The model had an overall accuracy of 0.75 for in directly classifying the three categories of benign, malignant and borderline SOTs, and a sensitivity of 0.89 for malignant, which was better than the overall diagnostic accuracy of 0.67 and sensitivity of 0.75 for malignant of the senior ultrasonographer.

Conclusion

DCNN model analysis of US images can provide complementary clinical diagnostic information and is thus a promising technique for effective differentiation of benign, borderline, and malignant SOTs.

SUBMITTER: Wang H

PROVIDER: S-EPMC8720926 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectivesTo develop and test the performance of computerized ultrasound image analysis using deep neural networks (DNNs) in discriminating between benign and malignant ovarian tumors and to compare its diagnostic accuracy with that of subjective assessment (SA) by an ultrasound expert.MethodsWe included 3077 (grayscale, n = 1927; power Doppler, n = 1150) ultrasound images from 758 women with ovarian tumors, who were classified prospectively by expert ultrasound examiners according to IOTA (International Ovarian Tumor Analysis) terms and definitions. Histological outcome from surgery (n = 634) or long-term (≥ 3 years) follow-up (n = 124) served as the gold standard. The dataset was split into a training set (n = 508; 314 benign and 194 malignant), a validation set (n = 100; 60 benign and 40 malignant) and a test set (n = 150; 75 benign and 75 malignant). We used transfer learning on three pre-trained DNNs: VGG16, ResNet50 and MobileNet. Each model was trained, and the outputs calibrated, using temperature scaling. An ensemble of the three models was then used to estimate the probability of malignancy based on all images from a given case. The DNN ensemble classified the tumors as benign or malignant (Ovry-Dx1 model); or as benign, inconclusive or malignant (Ovry-Dx2 model). The diagnostic performance of the DNN models, in terms of sensitivity and specificity, was compared to that of SA for classifying ovarian tumors in the test set.ResultsAt a sensitivity of 96.0%, Ovry-Dx1 had a specificity similar to that of SA (86.7% vs 88.0%; P = 1.0). Ovry-Dx2 had a sensitivity of 97.1% and a specificity of 93.7%, when designating 12.7% of the lesions as inconclusive. By complimenting Ovry-Dx2 with SA in inconclusive cases, the overall sensitivity (96.0%) and specificity (89.3%) were not significantly different from using SA in all cases (P = 1.0).ConclusionUltrasound image analysis using DNNs can predict ovarian malignancy with a diagnostic accuracy comparable to that of human expert examiners, indicating that these models may have a role in the triage of women with an ovarian tumor. © 2020 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Project description:ObjectivesThe use of magnetic resonance (MR) imaging in differentiation between benign and malignant adnexal masses in children and adolescents might be of great value in the diagnostic workup of sonographically indeterminate masses, since preserving fertility is of particular importance in this population. This systematic review evaluates the diagnostic value of MR imaging in children with an ovarian mass.MethodsThe review was made according to the PRISMA Statement. PubMed and EMBASE were systematically searched for studies on the use of MR imaging in differential diagnosis of ovarian masses in both adult women and children from 2008 to 2018.ResultsSixteen paediatric and 18 adult studies were included. In the included studies, MR imaging has shown good diagnostic performance in differentiating between benign and malignant ovarian masses. MR imaging techniques including diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) imaging seem to further improve the diagnostic performance.ConclusionThe addition of DWI with apparent diffusion coefficient (ADC) values measured in enhancing components of solid lesions and DCE imaging may further increase the good diagnostic performance of MR imaging in the pre-operative differentiation between benign and malignant ovarian masses by increasing specificity. Prospective age-specific studies are needed to confirm the high diagnostic performance of MR imaging in children and adolescents with a sonographically indeterminate ovarian mass.Key points• MR imaging, based on several morphological features, is of good diagnostic performance in differentiating between benign and malignant ovarian masses. Sensitivity and specificity varied between 84.8 to 100% and 20.0 to 98.4%, respectively. • MR imaging techniques like diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) imaging seem to improve the diagnostic performance. • Specific studies in children and adolescents with ovarian masses are required to confirm the suggested increased diagnostic performance of DWI and DCE in this population.

Project description:PurposeTo identify proteins and (molecular/biological) pathways associated with differences between benign and malignant epithelial ovarian tumors.Experimental proceduresSerum of six patients with a serous adenocarcinoma of the ovary was collected before treatment, with a control group consisting of six matched patients with a serous cystadenoma. In addition to the serum, homogeneous regions of cells exhibiting uniform histology were isolated from benign and cancerous tissue by laser microdissection. We subsequently employed label-free liquid chromatography tandem mass spectrometry (LC-MSe) to identify proteins in these serum and tissues samples. Analyses of differential expression between samples were performed using Bioconductor packages and in-house scripts in the statistical software package R. Hierarchical clustering and pathway enrichment analyses were performed, as well as network enrichment and interactome analysis using MetaCore.ResultsIn total, we identified 20 and 71 proteins that were significantly differentially expressed between benign and malignant serum and tissue samples, respectively. The differentially expressed protein sets in serum and tissue largely differed with only 2 proteins in common. MetaCore network analysis, however inferred GCR-alpha and Sp1 as common transcriptional regulators. Interactome analysis highlighted 14-3-3 zeta/delta, 14-3-3 beta/alpha, Alpha-actinin 4, HSP60, and PCBP1 as critical proteins in the tumor proteome signature based on their relative overconnectivity. The data have been deposited to the ProteomeXchange with identifier PXD001084.DiscussionOur analysis identified proteins with both novel and previously known associations to ovarian cancer biology. Despite the small overlap between differentially expressed protein sets in serum and tissue, APOA1 and Serotransferrin were significantly lower expressed in both serum and cancer tissue samples, suggesting a tissue-derived effect in serum. Pathway and subsequent interactome analysis also highlighted common regulators in serum and tissue samples, suggesting a yet unknown role for PCBP1 in ovarian cancer pathophysiology.