Search

Purpose: To compare the diagnostic accuracy of prone F-FDG PET/CT with that of contrast-enhanced MRI (CE-MRI) at 3 T in suspicious breast lesions. To evaluate the influence of tumour size on diagnostic accuracy and the use of maximum standardized uptake value (SUV) thresholds to differentiate malignant from benign breast lesions. Methods: A total of 172 consecutive patients with an imaging abnormality were included in this IRB-approved prospective study. All patients underwent F-FDG PET/CT and CE-MRI of the breast at 3 T in the prone position. Two reader teams independently evaluated the likelihood of malignancy as determined by F-FDG PET/CT and CE-MRI independently. F-FDG PET/CT data were qualitatively evaluated by visual interpretation. Quantitative assessment was performed by calculation of SUV. Sensitivity, specificity, diagnostic accuracy, area under the curve and interreader agreement were calculated for all lesions and for lesions <10 mm. Histopathology was used as the standard of reference. Results: There were 132 malignant and 40 benign lesions; 23 lesions (13.4 %) were <10 mm. Both F-FDG PET/CT and CE-MRI achieved an overall diagnostic accuracy of 93 %. There were no significant differences in sensitivity ( p = 0.125), specificity ( p = 0.344) or diagnostic accuracy ( p = 1). For lesions <10 mm, diagnostic accuracy deteriorated to 91 % with both F-FDG PET/CT and CE-MRI. Although no significant difference was found for lesions <10 mm, CE-MRI at 3 T seemed to be more sensitive but less specific than F-FDG PET/CT. Interreader agreement was excellent ( κ = 0.85 and κ = 0.92). SUV threshold was not helpful in differentiating benign from malignant lesions. Conclusion: F-FDG PET/CT and CE-MRI at 3 T showed equal diagnostic accuracies in breast cancer diagnosis. For lesions <10 mm, diagnostic accuracy deteriorated, but was equal for F-FDG PET/CT and CE-MRI at 3 T. For lesions <10 mm, CE-MRI at 3 T seemed to be more sensitive but less specific than F-FDG PET/CT. Quantitative assessment using an SUV threshold for differentiating benign from malignant lesions was not helpful in breast cancer diagnosis. [ABSTRACT FROM AUTHOR]

Simple Summary: Breast cancer is the second most common diagnosed malignancy in women worldwide. In this study, we examine the feasibility of breast tumor characterization based on [18F]FDG-PET/CT images using machine learning (ML) approaches in combination with data-preprocessing techniques. ML prediction models for breast cancer detection and the identification of breast cancer receptor status, proliferation rate, and molecular subtypes were established and evaluated. Furthermore, the importance of most repeatable features was investigated. Results displayed high performance of malignant/benign tumor differentiation and triple negative tumor subtype ML models. We observed high repeatability of radiomic features for both high performing predictive models. Background: This study investigated the performance of ensemble learning holomic models for the detection of breast cancer, receptor status, proliferation rate, and molecular subtypes from [18F]FDG-PET/CT images with and without incorporating data pre-processing algorithms. Additionally, machine learning (ML) models were compared with conventional data analysis using standard uptake value lesion classification. Methods: A cohort of 170 patients with 173 breast cancer tumors (132 malignant, 38 benign) was examined with [18F]FDG-PET/CT. Breast tumors were segmented and radiomic features were extracted following the imaging biomarker standardization initiative (IBSI) guidelines combined with optimized feature extraction. Ensemble learning including five supervised ML algorithms was utilized in a 100-fold Monte Carlo (MC) cross-validation scheme. Data pre-processing methods were incorporated prior to machine learning, including outlier and borderline noisy sample detection, feature selection, and class imbalance correction. Feature importance in each model was assessed by calculating feature occurrence by the R-squared method across MC folds. Results: Cross validation demonstrated high performance of the cancer detection model (80% sensitivity, 78% specificity, 80% accuracy, 0.81 area under the curve (AUC)), and of the triple negative tumor identification model (85% sensitivity, 78% specificity, 82% accuracy, 0.82 AUC). The individual receptor status and luminal A/B subtype models yielded low performance (0.46–0.68 AUC). SUVmax model yielded 0.76 AUC in cancer detection and 0.70 AUC in predicting triple negative subtype. Conclusions: Predictive models based on [18F]FDG-PET/CT images in combination with advanced data pre-processing steps aid in breast cancer diagnosis and in ML-based prediction of the aggressive triple negative breast cancer subtype. [ABSTRACT FROM AUTHOR]