Discriminating Between Benign and Malignant Solid Ovarian Tumors Based on Clinical and Radiomic Features of MRI

Multimodal ultrasound-based radiomics and deep learning for differential diagnosis of O-RADS 4-5 adnexal masses

BACKGROUND

Accurate differentiation between benign and malignant adnexal masses is crucial for patients to avoid unnecessary surgical interventions. Ultrasound (US) is the most widely utilized diagnostic and screening tool for gynecological diseases, with contrast-enhanced US (CEUS) offering enhanced diagnostic precision by clearly delineating blood flow within lesions. According to the Ovarian and Adnexal Reporting and Data System (O-RADS), masses classified as categories 4 and 5 carry the highest risk of malignancy. However, the diagnostic accuracy of US remains heavily reliant on the expertise and subjective interpretation of radiologists. Radiomics has demonstrated significant value in tumor differential diagnosis by extracting microscopic information imperceptible to the human eye. Despite this, no studies to date have explored the application of CEUS-based radiomics for differentiating adnexal masses. This study aims to develop and validate a multimodal US-based nomogram that integrates clinical variables, radiomics, and deep learning (DL) features to effectively distinguish adnexal masses classified as O-RADS 4-5.

METHODS

From November 2020 to March 2024, we enrolled 340 patients who underwent two-dimensional US (2DUS) and CEUS and had masses categorized as O-RADS 4-5. These patients were randomly divided into a training cohort and a test cohort in a 7:3 ratio. Adnexal masses were manually segmented from 2DUS and CEUS images. Using machine learning (ML) and DL techniques, five models were developed and validated to differentiate adnexal masses. The diagnostic performance of these models was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, specificity, precision, and F1-score. Additionally, a nomogram was constructed to visualize outcome measures.

RESULTS

The CEUS-based radiomics model outperformed the 2DUS model (AUC: 0.826 vs. 0.737). Similarly, the CEUS-based DL model surpassed the 2DUS model (AUC: 0.823 vs. 0.793). The ensemble model combining clinical variables, radiomics, and DL features achieved the highest AUC (0.929).

CONCLUSIONS

Our study confirms the effectiveness of CEUS-based radiomics for distinguishing adnexal masses with high accuracy and specificity using a multimodal US-based radiomics DL nomogram. This approach holds significant promise for improving the diagnostic precision of adnexal masses classified as O-RADS 4-5.

Advancing personalised care in ovarian cancer using CT and MRI radiomics

Radiomics, utilising quantitative feature extraction from CT and MR imaging, offers significant potential in advancing the diagnosis and management of ovarian cancer. Through the analysis of high-dimensional imaging data, radiomics models may capture subtle phenotypic variations in tumour heterogeneity, texture, and shape that extend beyond the capabilities of traditional imaging interpretation. CT-based radiomics excels in evaluating the prognostic significance of peritoneal disease dissemination and treatment response, while MRI-based models provide enhanced soft tissue characterisation, particularly in assessing tumour microstructure, vascularity, and cellularity. Studies demonstrate that these models can improve diagnostic accuracy, predict therapeutic outcomes and assist in risk stratification. However, standardisation of imaging acquisition protocols, feature extraction techniques and validation across diverse patient cohorts remains a challenge for the incorporation of radiomics into routine clinical practice. Evidence strongly supports the incorporation of radiomic features with molecular, genomic and clinical data in developing high-performance integrated radiomics models, which can facilitate precision oncology in ovarian cancer.

A clinicoradiological model based on clinical and CT features for preoperative prediction of histological classification in patients with epithelial ovarian cancers: a two-center study

OBJECTIVES

To develop and validate a clinicoradiological model integrating clinical and computed tomography (CT) features to preoperative predict histological classification in patients with epithelial ovarian cancers (EOCs).

METHODS

This retrospective study included 470 patients who were pathologically proven EOCs and performed by contrast enhanced CT before treatment from center I (training cohort, N = 329; internal test cohort, N = 141) and 83 EOC patients who were included as an external test cohort from center II. The univariate analysis and multivariate logistic regression analysis were used to select significant clinical and CT features. The significant clinical model was developed based on clinical characteristics. The significant radiological model was established by CT features. The significant clinical and CT features were used to construct the clinicoradiological model. Model performances were evaluated using the area under the receiver operating characteristic curve (AUC), calibration curve, the Brier score and decision curve analysis (DCA). The AUCs were compared by net reclassification index (NRI) and integrated discrimination improvement (IDI).

RESULTS

The significant clinical and CT parameters including age, transverse diameter, morphology, margin, ascites and lymphadenopathy were incorporated to build the clinicoradioligical model. The clinicoradiological model showed relatively satisfactory discrimination between type I and type II EOCs with the AUC of 0.841 (95% confidence interval [CI] 0.797-0.886), 0.874 (95% CI 0.811-0.937) and 0.826 (95% CI 0.729-0.923) in the training, internal and external test cohorts, respectively. The NRI and IDI showed the clinicoradiological model significantly performed than those of the clinical model (all P < 0.05). No statistical significance was found between radiological and clinicoradiological model. The clinicoradiological model demonstrated optimal classification accuracy and clinical application value.

CONCLUSION

The easily accessible nomogram based on the clinicoradiologic model showed favorable performance in distinguishing between type I and type II EOCs and could therefore be used to improve the clinical management of EOC patients.

The value of MRI in differentiating ovarian clear cell carcinoma from other adnexal masses with O-RADS MRI scores of 4-5

OBJECTIVE

To assess the utility of clinical and MRI features in distinguishing ovarian clear cell carcinoma (CCC) from adnexal masses with ovarian-adnexal reporting and data system (O-RADS) MRI scores of 4-5.

METHODS

This retrospective study included 850 patients with indeterminate adnexal masses on ultrasound. Two radiologists evaluated all preoperative MRIs using the O-RADS MRI risk stratification system. Patients with O-RADS MRI scores of 4-5 were divided into a training set (n = 135, hospital A) and a test set (n = 86, hospital B). Clinical and MRI features were compared between CCC and non-CCC patients. Analysis of variance and support vector machine were used to develop four CCC prediction models. Tenfold cross-validation was applied to determine the hyperparameters. Model performance was evaluated by the area under the curve (AUC) and decision curve.

RESULTS

221 patients were included (30 CCCs, 191 non-CCCs). CA125, HE4, CEA, ROMA, endometriosis, shape, parity, unilocular, component, the growth pattern of mural nodules, high signal on T1WI, number of nodules, the ratio of signal intensity, and the ADC value were significantly different between CCCs and non-CCCs. The kappa and interobserver correlation coefficient of each MRI feature exceeded 0.85. The comprehensive model combining clinical and MRI features had a greater AUC than the clinical model and tumour maker model (0.92 vs 0.66 and 0.78 in the test set; both p < 0.05), displaying improved net benefit.

CONCLUSIONS

The comprehensive model combining clinical and MRI features can effectively differentiate CCC from adnexal masses with O-RADS MRI scores of 4-5.

CRITICAL RELEVANCE STATEMENT

CCC has a high incidence rate in Asians and has limited sensitivity to platinum chemotherapy. This comprehensive model improves CCC prediction ability and clinical applicability for facilitating individualised clinical decision-making.

KEY POINTS

Identifying ovarian CCC preoperatively is beneficial for treatment planning. Ovarian CCC tends to be high-signal on T1WI, unilocular, big size, with endometriosis and low CEA. This model, integrating clinical and MRI features, can differentiate ovarian CCC from adnexal masses with O-RADS MRI scores 4-5.

Radiomics in radiology: What the radiologist needs to know about technical aspects and clinical impact

Radiomics represents the science of extracting and analyzing a multitude of quantitative features from medical imaging, revealing the quantitative potential of radiologic images. This scientific review aims to provide radiologists with a comprehensive understanding of radiomics, emphasizing its principles, applications, challenges, limits, and prospects. The limitations of standardization in current scientific production are analyzed, along with possible solutions proposed by some of the referenced papers. As the continuous evolution of medical imaging is ongoing, radiologists must be aware of new perspectives to play a central role in patient management.

CT-based radiomic analysis for categorization of ovarian sex cord-stromal tumors and epithelial ovarian cancers

PURPOSE

To evaluate the diagnostic potential of radiomic analyses based on machine learning that rely on contrast-enhanced computerized tomography (CT) for categorizing ovarian sex cord-stromal tumors (SCSTs) and epithelial ovarian cancers (EOCs).

METHODS

We included a total of 225 patients with 230 tumors, who were randomly divided into training and test cohorts with a ratio of 8:2. Radiomic features were extracted from each tumor and dimensionally reduced using LASSO. We used univariate and multivariate analyses to identify independent predictors from clinical features and conventional CT parameters. Clinic-radiological model, radiomics model and mixed model were constructed respectively. We evaluated model performance via analysis of the receiver operating characteristic (ROC) curve and area under ROC curves (AUCs), and compared it across models using the Delong test.

RESULTS

We selected a support vector machine as the best classifier. Both radiomic and mixed model achieved good classification accuracy with AUC values of 0.923/0.930 in the training cohort, and 0.879/0.909 in the test cohort. The mixed model performed significantly better than the model based on clinical radiological information, with AUC values of 0.930 versus 0.826 (p = 0.000) in the training cohort and 0.905 versus 0.788 (p = 0.042) in the test cohort.

CONCLUSION

Radiomic analysis based on CT images is a reliable and noninvasive tool for identifying SCSTs and EOCs, outperforming experience radiologists.

18F-Fluoro-2-Deoxyglucose Positron Emission Tomography/Computed Tomography Measures of Spatial Heterogeneity for Predicting Platinum Resistance of High-Grade Serous Ovarian Cancer

BACKGROUND

The purpose of this study is to construct models for predicting platinum resistance in high-grade serous ovarian cancer (HGSOC) derived from quantitative spatial heterogeneity indicators obtained from 18F-FDG PET/CT images.

METHODS

A retrospective study was conducted on patients diagnosed with HGSOC. Quantitative indicators of spatial heterogeneity were generated using conventional features and Haralick texture features from both CT and PET images. Three groups of predictive models (conventional, heterogeneity, and integrated) were built. Each group's optimal model was the one with the highest area under curve (AUC). Postoperative immunohistochemical staining for Ki-67 and p53 was conducted. The correlation between the heterogeneity indicators and scores for Ki-67 and p53 was assessed by Spearman's correlation coefficient (ρ).

RESULTS

A total of 286 patients (54.6 ± 9.3 years) were enrolled. And 107 spatial heterogeneity indicators were extracted. The optimal models for each group were obtained using the Gradient Boosting Machine (GBM) algorithm. There was an AUC of 0.790 (95% CI: 0.696, 0.885) in the conventional model for the validation set, and an AUC of 0.904 (95% CI: 0.842, 0.966) in the heterogeneity model for the validation set. The integrated model achieved the highest predictive performance, with an AUC value of 0.928 (95% CI: 0.872, 0.984) for the validation set. Spearman's correlation showed that HU_Kurtosis had the strongest correlation with p53 scores with ρ = 0.718, while cluster site entropy had the strongest correlation with Ki-67 scores with ρ = 0.753.

CONCLUSIONS

Adding quantitative spatial heterogeneity indicators derived from PET/CT images can improve the prediction of platinum resistance in patients with HGSOC. Spatial heterogeneity indicators were related to Ki-67 and p53 scores.

Ovarian cancer beyond imaging: integration of AI and multiomics biomarkers

High-grade serous ovarian cancer is the most lethal gynaecological malignancy. Detailed molecular studies have revealed marked intra-patient heterogeneity at the tumour microenvironment level, likely contributing to poor prognosis. Despite large quantities of clinical, molecular and imaging data on ovarian cancer being accumulated worldwide and the rise of high-throughput computing, data frequently remain siloed and are thus inaccessible for integrated analyses. Only a minority of studies on ovarian cancer have set out to harness artificial intelligence (AI) for the integration of multiomics data and for developing powerful algorithms that capture the characteristics of ovarian cancer at multiple scales and levels. Clinical data, serum markers, and imaging data were most frequently used, followed by genomics and transcriptomics. The current literature proves that integrative multiomics approaches outperform models based on single data types and indicates that imaging can be used for the longitudinal tracking of tumour heterogeneity in space and potentially over time. This review presents an overview of studies that integrated two or more data types to develop AI-based classifiers or prediction models.Relevance statement Integrative multiomics models for ovarian cancer outperform models using single data types for classification, prognostication, and predictive tasks.Key points• This review presents studies using multiomics and artificial intelligence in ovarian cancer.• Current literature proves that integrative multiomics outperform models using single data types.• Around 60% of studies used a combination of imaging with clinical data.• The combination of genomics and transcriptomics with imaging data was infrequently used.

A systematic review and meta-analysis of CT and MRI radiomics in ovarian cancer: methodological issues and clinical utility

OBJECTIVES

We aimed to present the state of the art of CT- and MRI-based radiomics in the context of ovarian cancer (OC), with a focus on the methodological quality of these studies and the clinical utility of these proposed radiomics models.

METHODS

Original articles investigating radiomics in OC published in PubMed, Embase, Web of Science, and the Cochrane Library between January 1, 2002, and January 6, 2023, were extracted. The methodological quality was evaluated using the radiomics quality score (RQS) and Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). Pairwise correlation analyses were performed to compare the methodological quality, baseline information, and performance metrics. Additional meta-analyses of studies exploring differential diagnoses and prognostic prediction in patients with OC were performed separately.

RESULTS

Fifty-seven studies encompassing 11,693 patients were included. The mean RQS was 30.7% (range - 4 to 22); less than 25% of studies had a high risk of bias and applicability concerns in each domain of QUADAS-2. A high RQS was significantly associated with a low QUADAS-2 risk and recent publication year. Significantly higher performance metrics were observed in studies examining differential diagnosis; 16 such studies as well as 13 exploring prognostic prediction were included in a separate meta-analysis, which revealed diagnostic odds ratios of 25.76 (95% confidence interval (CI) 13.50-49.13) and 12.55 (95% CI 8.38-18.77), respectively.

CONCLUSION

Current evidence suggests that the methodological quality of OC-related radiomics studies is unsatisfactory. Radiomics analysis based on CT and MRI showed promising results in terms of differential diagnosis and prognostic prediction.

CRITICAL RELEVANCE STATEMENT

Radiomics analysis has potential clinical utility; however, shortcomings persist in existing studies in terms of reproducibility. We suggest that future radiomics studies should be more standardized to better bridge the gap between concepts and clinical applications.

Artificial intelligence performance in image-based ovarian cancer identification: A systematic review and meta-analysis

BACKGROUND

Accurate identification of ovarian cancer (OC) is of paramount importance in clinical treatment success. Artificial intelligence (AI) is a potentially reliable assistant for the medical imaging recognition. We systematically review articles on the diagnostic performance of AI in OC from medical imaging for the first time.

METHODS

The Medline, Embase, IEEE, PubMed, Web of Science, and the Cochrane library databases were searched for related studies published until August 1, 2022. Inclusion criteria were studies that developed or used AI algorithms in the diagnosis of OC from medical images. The binary diagnostic accuracy data were extracted to derive the outcomes of interest: sensitivity (SE), specificity (SP), and Area Under the Curve (AUC). The study was registered with the PROSPERO, CRD42022324611.

FINDINGS

Thirty-four eligible studies were identified, of which twenty-eight studies were included in the meta-analysis with a pooled SE of 88% (95%CI: 85-90%), SP of 85% (82-88%), and AUC of 0.93 (0.91-0.95). Analysis for different algorithms revealed a pooled SE of 89% (85-92%) and SP of 88% (82-92%) for machine learning; and a pooled SE of 88% (84-91%) and SP of 84% (80-87%) for deep learning. Acceptable diagnostic performance was demonstrated in subgroup analyses stratified by imaging modalities (Ultrasound, Magnetic Resonance Imaging, or Computed Tomography), sample size (≤300 or >300), AI algorithms versus clinicians, year of publication (before or after 2020), geographical distribution (Asia or non Asia), and the different risk of bias levels (≥3 domain low risk or < 3 domain low risk).

INTERPRETATION

AI algorithms exhibited favorable performance for the diagnosis of OC through medical imaging. More rigorous reporting standards that address specific challenges of AI research could improve future studies.

FUNDING

This work was supported by the Natural Science Foundation of China (No. 82073647 to Q-JW and No. 82103914 to T-TG), LiaoNing Revitalization Talents Program (No. XLYC1907102 to Q-JW), and 345 Talent Project of Shengjing Hospital of China Medical University (No. M0268 to Q-JW and No. M0952 to T-TG).