A multicentric study of radiomics and artificial intelligence analysis on contrast-enhanced mammography to identify different histotypes of breast cancer.

OBJECTIVE: To evaluate the performance of radiomic analysis on contrast-enhanced mammography images to identify different histotypes of breast cancer mainly in order to predict grading, to identify hormone receptors, to discriminate human epidermal growth factor receptor 2 (HER2) and to identify luminal histotype of the breast cancer. METHODS: From four Italian centers were recruited 180 malignant lesions and 68 benign lesions. However, only the malignant lesions were considered for the analysis. All patients underwent contrast-enhanced mammography in cranium caudal (CC) and medium lateral oblique (MLO) view. Considering histological findings as the ground truth, four outcomes were considered: (1) G1 + G2 vs. G3; (2) HER2 + vs. HER2 - ; (3) HR + vs. HR - ; and (4) non-luminal vs. luminal A or HR + /HER2- and luminal B or HR + /HER2 + . For multivariate analysis feature selection, balancing techniques and patter recognition approaches were considered. RESULTS: The univariate findings showed that the diagnostic performance is low for each outcome, while the results of the multivariate analysis showed that better performances can be obtained. In the HER2 + detection, the best performance (73% of accuracy and AUC = 0.77) was obtained using a linear regression model (LRM) with 12 features extracted by MLO view. In the HR + detection, the best performance (77% of accuracy and AUC = 0.80) was obtained using a LRM with 14 features extracted by MLO view. In grading classification, the best performance was obtained by a decision tree trained with three predictors extracted by MLO view reaching an accuracy of 82% on validation set. In the luminal versus non-luminal histotype classification, the best performance was obtained by a bagged tree trained with 15 predictors extracted by CC view reaching an accuracy of 94% on validation set. CONCLUSIONS: The results suggest that radiomics analysis can be effectively applied to design a tool to support physician decision making in breast cancer classification. In particular, the classification of luminal versus non-luminal histotypes can be performed with high accuracy.

Radiomic and Artificial Intelligence Analysis with Textural Metrics Extracted by Contrast-Enhanced Mammography and Dynamic Contrast Magnetic Resonance Imaging to Detect Breast Malignant Lesions.

Purpose:The purpose of this study was to discriminate between benign and malignant breast lesions through several classifiers using, as predictors, radiomic metrics extracted from CEM and DCE-MRI images. In order to optimize the analysis, balancing and feature selection procedures were performed. Methods: Fifty-four patients with 79 histo-pathologically proven breast lesions (48 malignant lesions and 31 benign lesions) underwent both CEM and DCE-MRI. The lesions were retrospectively analyzed with radiomic and artificial intelligence approaches. Forty-eight textural metrics were extracted, and univariate and multivariate analyses were performed: non-parametric statistical test, receiver operating characteristic (ROC) and machine learning classifiers. Results: Considering the single metrics extracted from CEM, the best predictors were KURTOSIS (area under ROC curve (AUC) = 0.71) and SKEWNESS (AUC = 0.71) calculated on late MLO view. Considering the features calculated from DCE-MRI, the best predictors were RANGE (AUC = 0.72), ENERGY (AUC = 0.72), ENTROPY (AUC = 0.70) and GLN (gray-level nonuniformity) of the gray-level run-length matrix (AUC = 0.72). Considering the analysis with classifiers and an unbalanced dataset, no significant results were obtained. After the balancing and feature selection procedures, higher values of accuracy, specificity and AUC were reached. The best performance was obtained considering 18 robust features among all metrics derived from CEM and DCE-MRI, using a linear discriminant analysis (accuracy of 0.84 and AUC = 0.88). Conclusions: Classifiers, adjusted with adaptive synthetic sampling and feature selection, allowed for increased diagnostic performance of CEM and DCE-MRI in the differentiation between benign and malignant lesions.

Radiomics and Artificial Intelligence Analysis with Textural Metrics Extracted by Contrast-Enhanced Mammography in the Breast Lesions Classification.

The aim of the study was to estimate the diagnostic accuracy of textural features extracted by dual-energy contrast-enhanced mammography (CEM) images, by carrying out univariate and multivariate statistical analyses including artificial intelligence approaches. In total, 80 patients with known breast lesion were enrolled in this prospective study according to regulations issued by the local Institutional Review Board. All patients underwent dual-energy CEM examination in both craniocaudally (CC) and double acquisition of mediolateral oblique (MLO) projections (early and late). The reference standard was pathology from a surgical specimen for malignant lesions and pathology from a surgical specimen or fine needle aspiration cytology, core or Tru-Cut needle biopsy, and vacuum assisted breast biopsy for benign lesions. In total, 104 samples of 80 patients were analyzed. Furthermore, 48 textural parameters were extracted by manually segmenting regions of interest. Univariate and multivariate approaches were performed: non-parametric Wilcoxon-Mann-Whitney test; receiver operating characteristic (ROC), linear classifier (LDA), decision tree (DT), k-nearest neighbors (KNN), artificial neural network (NNET), and support vector machine (SVM) were utilized. A balancing approach and feature selection methods were used. The univariate analysis showed low accuracy and area under the curve (AUC) for all considered features. Instead, in the multivariate textural analysis, the best performance considering the CC view (accuracy (ACC) = 0.75; AUC = 0.82) was reached with a DT trained with leave-one-out cross-variation (LOOCV) and balanced data (with adaptive synthetic (ADASYN) function) and a subset of three robust textural features (MAD, VARIANCE, and LRLGE). The best performance (ACC = 0.77; AUC = 0.83) considering the early-MLO view was reached with a NNET trained with LOOCV and balanced data (with ADASYN function) and a subset of ten robust features (MEAN, MAD, RANGE, IQR, VARIANCE, CORRELATION, RLV, COARSNESS, BUSYNESS, and STRENGTH). The best performance (ACC = 0.73; AUC = 0.82) considering the late-MLO view was reached with a NNET trained with LOOCV and balanced data (with ADASYN function) and a subset of eleven robust features (MODE, MEDIAN, RANGE, RLN, LRLGE, RLV, LZLGE, GLV_GLSZM, ZSV, COARSNESS, and BUSYNESS). Multivariate analyses using pattern recognition approaches, considering 144 textural features extracted from all three mammographic projections (CC, early MLO, and late MLO), optimized by adaptive synthetic sampling and feature selection operations obtained the best results (ACC = 0.87; AUC = 0.90) and showed the best performance in the discrimination of benign and malignant lesions.

Evaluation of average glandular dose and investigation of the relationship with compressed breast thickness in dual energy contrast enhanced digital mammography and digital breast tomosynthesis.

PURPOSE: To quantitatively assess the dose of Dual energy contrast enhanced digital mammography (CEDM) and digital breast tomosynthesis (DBT) and to investigate the relationship between average absorbed glandular dose (AGD), compressed breast thickness (CBT) and compression force (CF). MATERIALS AND METHODS: All CEDM and DBT examinations were performed in cranio-caudal (CC) and medio-lateral oblique (MLO) view. Exposure parameters of 135 mammographic procedures that using AEC (automatic exposure control) mode were recorded. AGDs were calculated. Kruskal Wallis test was performed. RESULTS: CBT population ranged from 23 to 94 mm with a thickness median value of 52 mm in CC view and of 57 mm in MLO views. CEDM AGD median value was significatively lower than DBT AGD in each views (p << 0.01). AGD showed a positive correlation and linear regression with CBT for both CEDM and DBT while CF did not show a correlation and linear regression with AGD. The highest values were found for MLO view: R2 of 0.74 for CEDM and R2 of 0.61 for DBT. Kruskal Wallis test shows that there was a difference statistically significant between AGD values of CEDM and DBT in CC view respect to MLO views (p < 0.01). CONCLUSIONS: Dose values of both techniques meet the recommendations for maximum dose in mammography. The results of the present study indicated that there was significant difference between AGD for CEDM and DBT exposure in different views (AGD in CC views had the lowest value) and that CBT could influence the AGD while CF was not correlated to AGD.

Abbreviated breast dynamic contrast-enhanced MR imaging for lesion detection and characterization: the experience of an Italian oncologic center.

BACKGROUND: To evaluate the performance of an abbreviated dynamic contrast-enhanced MR imaging (MRI) protocol for breast cancer detection; a comparison with the complete diagnostic protocol has been conducted. METHODS: A retrospective analysis on 508 patients was performed. Abbreviated protocol (AP) included one pre-contrast and the first post-contrast T1-weighted series. Complete protocol (CP) consisted of four post-contrast and one pre-contrast T1-weighted series. Diagnostic performance was assessed for AP and CP. Performance comparison was made using McNemar's test for sensitivity and specificity and Moskowitz and Pepe's method as regards negative predictive value (NPV) and positive predictive value (PPV). AP has been realized in two different ways (AP1 and AP2) and they were compared by means of Cohen's κ. RESULTS: Both CP and AP revealed 206 of 207 cancers. There were no statistically significant differences between AP and CP diagnostic performance (P > 0.05). NPVs of CP and both versions of AP (99.57 vs. 99.56%, P = 0.39), as well as the specificity (77.08 vs. 75.42%, P = 0.18), were substantially equivalent. Relative predictive value method did not reveal the presence of a statistically significant difference between the PPV of CP and both versions of AP (74.91 vs. 73.57%, P = 0.099). Analysis for single lesion confirmed that both CP and AP had equivalent results: CP and AP revealed 280 of 281 malignancies. NPVs of CP and both AP versions, as well as the specificity (P > 0.05), were substantially equivalent. Relative predictive value method did not reveal the presence of a significant difference between the PPV of CP and both AP versions (70.89 vs. 70.18%, P = 0.25; 70.89 vs. 70.00%, P = 0.13). CONCLUSIONS: Abbreviated approach to breast MRI examination reduces the image acquisition and the reading time associated with MR substantially without influencing the diagnostic accuracy (high sensitivity and NPV >99.5%). AP could translate into cost-savings and could enable a higher number of examinations within the same MR session.

Surgical impact of preoperative breast MRI in women below 40 years of age.

Our aim was to evaluate the surgical impact of preoperative MRI in young patients. We reviewed a single-institution database of 283 consecutive patients below 40 years of age and who were treated for breast cancer. Thirty-seven (13 %) patients who received neoadjuvant chemotherapy were excluded. The remaining 246 patients included 124 (50 %) who preoperatively underwent conventional imaging (CI), i.e., mammography/ultrasonography (CI-group), and 122 (50 %) who underwent CI and dynamic MRI (CI + MRI-group). Pathology of surgical specimens served as a reference standard. Mann-Whitney, χ (2), and McNemar statistics were used. There were no significant differences between groups in terms of age, tumor pathologic subtype, stage, receptor, or nodal status. The mastectomy rate was 111/246 (45 %) overall but was significantly different between groups (46/124, 37 %, for the CI group and 65/122, 53 %, for the CI + MRI group; p = 0.011). Of 122 CI + MRI patients, 46 (38 %) would have undergone mastectomy due to CI alone, while MRI determined 19 additional mastectomies, increasing the mastectomy rate from 38 % to 53 % (p < 0.001). The number of patients with multifocal, multicentric, synchronous, or bilateral cancers was significantly different between groups (10/124, 8 %, for the CI group and 33/122, 27 %, for the CI + MRI group; p < 0.001). In the CI + MRI group, multifocal, multicentric, or synchronous bilateral cancers were detected with mammography in 5/33 (15 %) patients, with ultrasonography in 15/33 (45 %) patients, and with MRI in 32/33 (97 %) patients (p < 0.005). Two mastectomies were due to false positives at both conventional tests in the CI group (2/124, 1.6 %) and two mastectomies were due to MRI false positives in the CI + MRI group (2/122, 1.6 %). In conclusion, breast cancer in young patients was treated with mastectomy in 37-38 % of cases on the basis of CI only and in these patients MRI was more sensitive than CI for multifocal, multicentric, or synchronous bilateral cancers, resulting in an additional mastectomy rate of 15 %. A low probability of inappropriate imaging-based decision-making for mastectomy exists for both CI alone and for CI + MRI, making presurgical needle biopsy mandatory for findings that suggest a need for mastectomy.