Preoperative Prediction of Metastasis for Ovarian Cancer Based on Computed Tomography Radiomics Features and Clinical Factors

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.

Development and validation of radiomics nomogram for metastatic status of epithelial ovarian cancer

To develop and validate an enhanced CT-based radiomics nomogram for evaluating preoperative metastasis risk of epithelial ovarian cancer (EOC). One hundred and nine patients with histologically confirmed EOC were retrospectively enrolled. The volume of interest (VOI) was delineated in preoperative enhanced CT images, and 851 radiomics features were extracted. The radiomics features were selected by the least absolute shrinkage and selection operator (LASSO), and the rad-score was calculated using the formula of the radiomics label. A clinical model, radiomics model, and combined model were constructed using the logistic regression classification algorithm. Receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA) were used to evaluate the diagnostic performance of the models. Seventy-five patients (68.8%) were histologically confirmed to have metastasis. Eleven optimal radiomics features were retained by the LASSO algorithm to develop the radiomic model. The combined model for evaluating metastasis of EOC achieved area under the curve (AUC) values of 0.929 (95% CI 0.8593-0.9996) in the training cohort and 0.909 (95% CI 0.7921-1.0000) in the test cohort. To facilitate clinical use, a radiomic nomogram was built by combining the clinical characteristics with rad-score. The DCA indicated that the nomogram had the most significant net benefit when the threshold probability exceeded 15%, surpassing the benefits of both the treat-all and treat-none strategies. Compared with clinical model and radiomics model, the radiomics nomogram has the best diagnostic performance in evaluating EOC metastasis. The nomogram is a useful and convenient tool for clinical doctors to develop personalized treatment plans for EOC patients.

CT-Based Radiomics for the Preoperative Prediction of Occult Peritoneal Metastasis in Epithelial Ovarian Cancers

RATIONALE AND OBJECTIVES

The objective of this study was to develop a comprehensive combined model for predicting occult peritoneal metastasis (OPM) in epithelial ovarian cancers (EOCs) using radiomics features derived from computed tomography (CT) and clinical-radiological predictors.

MATERIALS AND METHODS

A total of 224 patients with EOCs were randomly divided into training dataset (N = 156) and test dataset (N = 86). Five clinical factors and seven radiological features were collected. The radiomics features were extracted from CT images of each patient. Multivariate logistic regression was employed to construct clinical and radiological models. The correlation analysis and least absolute shrinkage and selection operator algorithm were used to select radiomics features and build radiomics model. The important clinical, radiological factors, and radiomics features were integrated into a combined model by multivariate logistic regression. Receiver operating characteristics curve with area under the curve (AUC) were used to evaluate and compare predictive performance.

RESULTS

Carbohydrate antigen 125 (CA-125) and human epididymal protein 4 (HE-4) were independent clinical predictors. Laterality, thickened septa and margin were independent radiological predictors. In the training dataset, the AUCs for the clinical, radiological and radiomics models in evaluating OPM were 0.759, 0.819, and 0.830, respectively. In the test dataset, the AUCs for these models were 0.846, 0.835, and 0.779, respectively. The combined model outperformed other models in both the training and the test datasets with AUCs of 0.901 and 0.912, respectively. Decision curve analysis indicated that the combined model yielded a higher net benefit compared to the other models.

CONCLUSION

The combined model, integrating radiomics features with clinical and radiological predictors exhibited improved accuracy in predicting OPM in EOCs.

The radiomics-clinical nomogram for predicting the response to initial superselective arterial embolization in renal angiomyolipoma, a preliminary study

Purpose

The aim of this study was to explore a radiomics-clinical model for predicting the response to initial superselective arterial embolization (SAE) in renal angiomyolipoma (RAML).

Materials and methods

A total of 78 patients with RAML were retrospectively enrolled. Clinical data were recorded and evaluated. Radiomic features were extracted from preoperative contrast-enhanced CT (CECT). Least absolute shrinkage and selection operator (LASSO) and intra- and inter-class correlation coefficients (ICCs) were used in feature selection. Logistic regression analysis was performed to develop the radiomics, clinical, and combined models where the fivefold cross-validation method was used. The predictive performance and calibration were evaluated by the receiver operating characteristic (ROC) curve and calibration curve. Decision curve analysis (DCA) was used to measure clinical usefulness.

Results

The tumor shrinkage rate was 29.7% in total, and both fat and angiomyogenic components were significantly reduced. In the radiomics model, 12 significant features were selected. In the clinical model, maximum diameter (p = 0.001), angiomyogenic tissue ratio (p = 0.032), aneurysms (p = 0.048), and post-SAE time (p = 0.002) were significantly associated with greater volume reduction after SAE. Because of the severe linear dependence between radiomics signature and some clinical parameters, the combined model eventually included Rad-score, aneurysm, and post-SAE time. The radiomics-clinical model showed better discrimination (mean AUC = 0.83) than the radiomics model (mean AUC = 0.60) and the clinical model (mean AUC = 0.82). Calibration curve and DCA showed the goodness of fit and clinical usefulness of the radiomics-clinical model.

Conclusions

The radiomics-clinical model incorporating radiomics features and clinical parameters can potentially predict the positive response to initial SAE in RAML and provide support for clinical treatment decisions.

Advances in artificial intelligence for the diagnosis and treatment of ovarian cancer (Review)

Artificial intelligence (AI) has emerged as a crucial technique for extracting high‑throughput information from various sources, including medical images, pathological images, and genomics, transcriptomics, proteomics and metabolomics data. AI has been widely used in the field of diagnosis, for the differentiation of benign and malignant ovarian cancer (OC), and for prognostic assessment, with favorable results. Notably, AI‑based radiomics has proven to be a non‑invasive, convenient and economical approach, making it an essential asset in a gynecological setting. The present study reviews the application of AI in the diagnosis, differentiation and prognostic assessment of OC. It is suggested that AI‑based multi‑omics studies have the potential to improve the diagnostic and prognostic predictive ability in patients with OC, thereby facilitating the realization of precision medicine.

Development and validation of an ultrasound‑based radiomics nomogram to predict lymph node status in patients with high-grade serous ovarian cancer: a retrospective analysis

BACKGROUND

Despite advances in medical imaging technology, the accurate preoperative prediction of lymph node status remains challenging in ovarian cancer. This retrospective study aimed to investigate the feasibility of using ultrasound-based radiomics combined with preoperative clinical characteristics to predict lymph node metastasis (LNM) in patients with high-grade serous ovarian cancer (HGSOC).

RESULTS

Patients with 401 HGSOC lesions from two institutions were enrolled: institution 1 for the training cohort (n = 322) and institution 2 for the external test cohort (n = 79). Radiomics features were extracted from the three preoperative ultrasound images of each lesion. During feature selection, primary screening was first performed using the sample variance F-value, followed by recursive feature elimination (RFE) to filter out the 12 most significant features for predicting LNM. The radscore derived from these 12 radiomic features and three clinical characteristics were used to construct a combined model and nomogram to predict LNM, and subsequent 10-fold cross-validation was performed. In the test phase, the three models were tested with external test cohort. The radiomics model had an area under the curve (AUC) of 0.899 (95% confidence interval [CI]: 0.864-0.933) in the training cohort and 0.855 (95%CI: 0.774-0.935) in the test cohort. The combined model showed good calibration and discrimination in the training cohort (AUC = 0.930) and test cohort (AUC = 0.881), which were superior to those of the radiomic and clinical models alone.

CONCLUSIONS

The nomogram consisting of the radscore and preoperative clinical characteristics showed good diagnostic performance in predicting LNM in patients with HGSOC. It may be used as a noninvasive method for assessing the lymph node status in these patients.

Multitask prediction models for serous ovarian cancer by preoperative CT image assessments based on radiomics

Objective

High-grade serous ovarian cancer (HGSOC) has the highest mortality rate among female reproductive system tumors. Accurate preoperative assessment is crucial for treatment planning. This study aims to develop multitask prediction models for HGSOC using radiomics analysis based on preoperative CT images.

Methods

This study enrolled 112 patients diagnosed with HGSOC. Laboratory findings, including serum levels of CA125, HE-4, and NLR, were collected. Radiomic features were extracted from manually delineated ROI on CT images by two radiologists. Classification models were developed using selected optimal feature sets to predict R0 resection, lymph node invasion, and distant metastasis status. Model evaluation was conducted by quantifying receiver operating curves (ROC), calculating the area under the curve (AUC), De Long's test.

Results

The radiomics models applied to CT images demonstrated superior performance in the testing set compared to the clinical models. The area under the curve (AUC) values for the combined model in predicting R0 resection were 0.913 and 0.881 in the training and testing datasets, respectively. De Long's test indicated significant differences between the combined and clinical models in the testing set (p = 0.003). For predicting lymph node invasion, the AUCs of the combined model were 0.868 and 0.800 in the training and testing datasets, respectively. The results also revealed significant differences between the combined and clinical models in the testing set (p = 0.002). The combined model for predicting distant metastasis achieved AUCs of 0.872 and 0.796 in the training and test datasets, respectively. The combined model displayed excellent agreement between observed and predicted results in predicting R0 resection, while the radiomics model demonstrated better calibration than both the clinical model and combined model in predicting lymph node invasion and distant metastasis. The decision curve analysis (DCA) for predicting R0 resection favored the combined model over both the clinical and radiomics models, whereas for predicting lymph node invasion and distant metastasis, DCA favored the radiomics model over both the clinical model and combined model.

Conclusion

The identified radiomics signature holds potential value in preoperatively evaluating the R0, lymph node invasion and distant metastasis in patients with HGSC. The radiomics nomogram demonstrated the incremental value of clinical predictors for surgical outcome and metastasis estimation.

Diagnostic value of a CT-based radiomics nomogram for discrimination of benign and early stage malignant ovarian tumors

BACKGROUND

This study aimed to identify the diagnostic value of models constructed using computed tomography-based radiomics features for discrimination of benign and early stage malignant ovarian tumors.

METHODS

The imaging and clinicopathological data of 197 cases of benign and early stage malignant ovarian tumors (FIGO stage I/II), were retrospectively analyzed. The patients were randomly assigned into training data set and validation data set. Radiomics features were extracted from images of plain computed tomography scan and contrast-enhanced computed tomography scan, were then screened in the training data set, and a radiomics model was constructed. Multivariate logistic regression analysis was used to construct a radiomic nomogram, containing the traditional diagnostic model and the radiomics model. Moreover, the decision curve analysis was used to assess the clinical application value of the radiomics nomogram.

RESULTS

Six textural features with the greatest diagnostic efficiency were finally screened. The value of the area under the receiver operating characteristic curve showed that the radiomics nomogram was superior to the traditional diagnostic model and the radiomics model (P < 0.05) in the training data set. In the validation data set, the radiomics nomogram was superior to the traditional diagnostic model (P < 0.05), but there was no statistically significant difference compared to the radiomics model (P > 0.05). The calibration curve and the Hosmer-Lemeshow test revealed that the three models all had a great degree of fit (All P > 0.05). The results of decision curve analysis indicated that utilization of the radiomics nomogram to distinguish benign and early stage malignant ovarian tumors had a greater clinical application value when the risk threshold was 0.4-1.0.

CONCLUSIONS

The computed tomography-based radiomics nomogram could be a non-invasive and reliable imaging method to discriminate benign and early stage malignant ovarian tumors.

Radiomics in the evaluation of ovarian masses - a systematic review

OBJECTIVES

The study aim was to conduct a systematic review of the literature reporting the application of radiomics to imaging techniques in patients with ovarian lesions.

METHODS

MEDLINE/PubMed, Web of Science, Scopus, EMBASE, Ovid and ClinicalTrials.gov were searched for relevant articles. Using PRISMA criteria, data were extracted from short-listed studies. Validity and bias were assessed independently by 2 researchers in consensus using the Quality in Prognosis Studies (QUIPS) tool. Radiomic Quality Score (RQS) was utilised to assess radiomic methodology.

RESULTS

After duplicate removal, 63 articles were identified, of which 33 were eligible. Fifteen assessed lesion classifications, 10 treatment outcomes, 5 outcome predictions, 2 metastatic disease predictions and 1 classification/outcome prediction. The sample size ranged from 28 to 501 patients. Twelve studies investigated CT, 11 MRI, 4 ultrasound and 1 FDG PET-CT. Twenty-three studies (70%) incorporated 3D segmentation. Various modelling methods were used, most commonly LASSO (least absolute shrinkage and selection operator) (10/33). Five studies (15%) compared radiomic models to radiologist interpretation, all demonstrating superior performance. Only 6 studies (18%) included external validation. Five studies (15%) had a low overall risk of bias, 9 (27%) moderate, and 19 (58%) high risk of bias. The highest RQS achieved was 61.1%, and the lowest was - 16.7%.

CONCLUSION

Radiomics has the potential as a clinical diagnostic tool in patients with ovarian masses and may allow better lesion stratification, guiding more personalised patient care in the future. Standardisation of the feature extraction methodology, larger and more diverse patient cohorts and real-world evaluation is required before clinical translation.

CLINICAL RELEVANCE STATEMENT

Radiomics shows promising results in improving lesion stratification, treatment selection and outcome prediction. Modelling with larger cohorts and real-world evaluation is required before clinical translation.

KEY POINTS

• Radiomics is emerging as a tool for enhancing clinical decisions in patients with ovarian masses. • Radiomics shows promising results in improving lesion stratification, treatment selection and outcome prediction. • Modelling with larger cohorts and real-world evaluation is required before clinical translation.

Radiomics-Clinical AI Model with Probability Weighted Strategy for Prognosis Prediction in Non-Small Cell Lung Cancer

In this study, we propose a radiomics clinical probability-weighted model for the prediction of prognosis for non-small cell lung cancer (NSCLC). The model combines radiomics features extracted from radiotherapy (RT) planning images with clinical factors such as age, gender, histology, and tumor stage. CT images with radiotherapy structures of 422 NSCLC patients were retrieved from The Cancer Imaging Archive (TCIA). Radiomic features were extracted from gross tumor volumes (GTVs). Five machine learning algorithms, namely decision trees (DT), random forests (RF), extreme boost (EB), support vector machine (SVM) and generalized linear model (GLM) were optimized by a voted ensemble machine learning (VEML) model. A probabilistic weighted approach is used to incorporate the uncertainty associated with both radiomic and clinical features and to generate a probabilistic risk score for each patient. The performance of the model is evaluated using a receiver operating characteristic (ROC). The Radiomic model, clinical factor model, and combined radiomic clinical probability-weighted model demonstrated good performance in predicting NSCLC survival with AUC of 0.941, 0.856 and 0.949, respectively. The combined radiomics clinical probability-weighted enhanced model achieved significantly better performance than the radiomic model in 1-year survival prediction (chi-square test, p < 0.05). The proposed model has the potential to improve NSCLC prognosis and facilitate personalized treatment decisions.

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.

Radiomics and Radiogenomics of Ovarian Cancer

Ovarian cancer, one of the deadliest gynecologic malignancies, is characterized by high intra- and inter-site genomic and phenotypic heterogeneity. The traditional information provided by the conventional interpretation of diagnostic imaging studies cannot adequately represent this heterogeneity. Radiomics analyses can capture the complex patterns related to the microstructure of the tissues and provide quantitative information about them. This review outlines how radiomics and its integration with other quantitative biological information, like genomics and proteomics, can impact the clinical management of ovarian cancer.

CT-based machine learning radiomics predicts CCR5 expression level and survival in ovarian cancer

OBJECTIVE

We aimed to evaluate the prognostic value of C-C motif chemokine receptor type 5 (CCR5) expression level for patients with ovarian cancer and to establish a radiomics model that can predict CCR5 expression level using The Cancer Imaging Archive (TCIA) and The Cancer Genome Atlas (TCGA) database.

METHODS

A total of 343 cases of ovarian cancer from the TCGA were used for the gene-based prognostic analysis. Fifty seven cases had preoperative computed tomography (CT) images stored in TCIA with genomic data in TCGA were used for radiomics feature extraction and model construction. 89 cases with both TCGA and TCIA clinical data were used for radiomics model evaluation. After feature extraction, a radiomics signature was constructed using the least absolute shrinkage and selection operator (LASSO) regression analysis. A prognostic scoring system incorporating radiomics signature based on CCR5 expression level and clinicopathologic risk factors was proposed for survival prediction.

RESULTS

CCR5 was identified as a differentially expressed prognosis-related gene in tumor and normal sample, which were involved in the regulation of immune response and tumor invasion and metastasis. Four optimal radiomics features were selected to predict overall survival. The performance of the radiomics model for predicting the CCR5 expression level with 10-fold cross- validation achieved Area Under Curve (AUCs) of 0.770 and of 0.726, respectively, in the training and validation sets. A predictive nomogram was generated based on the total risk score of each patient, the AUCs of the time-dependent receiver operating characteristic (ROC) curve of the model was 0.8, 0.673 and 0.792 for 1-year, 3-year and 5-year, respectively. Along with clinical features, important imaging biomarkers could improve the overall survival accuracy of the prediction model.

CONCLUSION

The expression levels of CCR5 can affect the prognosis of patients with ovarian cancer. CT-based radiomics could serve as a new tool for prognosis prediction.

Computed Tomographic Radiomics in Differentiating Histologic Subtypes of Epithelial Ovarian Carcinoma

IMPORTANCE

Epithelial ovarian carcinoma is heterogeneous and classified according to the World Health Organization Tumour Classification, which is based on histologic features and molecular alterations. Preoperative prediction of the histologic subtypes could aid in clinical management and disease prognostication.

OBJECTIVE

To assess the value of radiomics based on contrast-enhanced computed tomography (CT) in differentiating histologic subtypes of epithelial ovarian carcinoma in multicenter data sets.

DESIGN, SETTING, AND PARTICIPANTS

In this diagnostic study, 665 patients with histologically confirmed epithelial ovarian carcinoma were retrospectively recruited from 4 centers (Hong Kong, Guangdong Province of China, and Seoul, South Korea) between January 1, 2012, and February 28, 2022. The patients were randomly divided into a training cohort (n = 532) and a testing cohort (n = 133) with a ratio of 8:2. This process was repeated 100 times. Tumor segmentation was manually delineated on each section of contrast-enhanced CT images to encompass the entire tumor. The Mann-Whitney U test and voted least absolute shrinkage and selection operator were performed for feature reduction and selection. Selected features were used to build the logistic regression model for differentiating high-grade serous carcinoma and non-high-grade serous carcinoma.

EXPOSURES

Contrast-enhanced CT-based radiomics.

MAIN OUTCOMES AND MEASURES

Intraobserver and interobserver reproducibility of tumor segmentation were measured by Dice similarity coefficients. The diagnostic efficiency of the model was assessed by receiver operating characteristic curve and area under the curve.

RESULTS

In this study, 665 female patients (mean [SD] age, 53.6 [10.9] years) with epithelial ovarian carcinoma were enrolled and analyzed. The Dice similarity coefficients of intraobserver and interobserver were all greater than 0.80. Twenty radiomic features were selected for modeling. The areas under the curve of the logistic regression model in differentiating high-grade serous carcinoma and non-high-grade serous carcinoma were 0.837 (95% CI, 0.835-0.838) for the training cohort and 0.836 (95% CI, 0.833-0.840) for the testing cohort.

CONCLUSIONS AND RELEVANCE

In this diagnostic study, radiomic features extracted from contrast-enhanced CT were useful in the classification of histologic subtypes in epithelial ovarian carcinoma. Intraobserver and interobserver reproducibility of tumor segmentation was excellent. The proposed logistic regression model offered excellent discriminative ability among histologic subtypes.

Risk Prediction of Second Primary Malignancies in Primary Early-Stage Ovarian Cancer Survivors: A SEER-Based National Population-Based Cohort Study

Purpose

This study aimed to characterize the clinical features of early-stage ovarian cancer (OC) survivors with second primary malignancies (SPMs) and provided a prediction tool for individualized risk of developing SPMs.

Methods

Data were obtained from the Surveillance, Epidemiology and End Results (SEER) database during 1998-2013. Considering non-SPM death as a competing event, the Fine and Gray model and the corresponding nomogram were used to identify the risk factors for SPMs and predict the SPM probabilities after the initial OC diagnosis. The decision curve analysis (DCA) was performed to evaluate the clinical utility of our proposed model.

Results

A total of 14,314 qualified patients were enrolled. The diagnosis rate and the cumulative incidence of SPMs were 7.9% and 13.6% [95% confidence interval (CI) = 13.5% to 13.6%], respectively, during the median follow-up of 8.6 years. The multivariable competing risk analysis suggested that older age at initial cancer diagnosis, white race, epithelial histologic subtypes of OC (serous, endometrioid, mucinous, and Brenner tumor), number of lymph nodes examined (<12), and radiotherapy were significantly associated with an elevated SPM risk. The DCA revealed that the net benefit obtained by our proposed model was higher than the all-screening or no-screening scenarios within a wide range of risk thresholds (1% to 23%).

Conclusion

The competing risk nomogram can be potentially helpful for assisting physicians in identifying patients with different risks of SPMs and scheduling risk-adapted clinical management. More comprehensive data on treatment regimens and patient characteristics may help improve the predictability of the risk model for SPMs.

CT-Based Radiomics and Deep Learning for BRCA Mutation and Progression-Free Survival Prediction in Ovarian Cancer Using a Multicentric Dataset

PURPOSE

Build predictive radiomic models for early relapse and BRCA mutation based on a multicentric database of high-grade serous ovarian cancer (HGSOC) and validate them in a test set coming from different institutions.

METHODS

Preoperative CTs of patients with HGSOC treated at four referral centers were retrospectively acquired and manually segmented. Hand-crafted features and deep radiomics features were extracted respectively by dedicated software (MODDICOM) and a dedicated convolutional neural network (CNN). Features were selected with and without prior harmonization (ComBat harmonization), and models were built using different machine learning algorithms, including clinical variables.

RESULTS

We included 218 patients. Radiomic models showed low performance in predicting both BRCA mutation (AUC in test set between 0.46 and 0.59) and 1-year relapse (AUC in test set between 0.46 and 0.56); deep learning models demonstrated similar results (AUC in the test of 0.48 for BRCA and 0.50 for relapse). The inclusion of clinical variables improved the performance of the radiomic models to predict BRCA mutation (AUC in the test set of 0.74).

CONCLUSIONS

In our multicentric dataset, representative of a real-life clinical scenario, we could not find a good radiomic predicting model for PFS and BRCA mutational status, with both traditional radiomics and deep learning, but the combination of clinical and radiomic models improved model performance for the prediction of BRCA mutation. These findings highlight the need for standardization through the whole radiomic pipelines and robust multicentric external validations of results.