A deep learning model for breast ductal carcinoma in situ classification in whole slide images

The pathological differential diagnosis between breast ductal carcinoma in situ (DCIS) and invasive ductal carcinoma (IDC) is of pivotal importance for determining optimum cancer treatment(s) and clinical outcomes. Since conventional diagnosis by pathologists using microscopes is limited in terms of human resources, it is necessary to develop new techniques that can rapidly and accurately diagnose large numbers of histopathological specimens. Computational pathology tools which can assist pathologists in detecting and classifying DCIS and IDC from whole slide images (WSIs) would be of great benefit for routine pathological diagnosis. In this paper, we trained deep learning models capable of classifying biopsy and surgical histopathological WSIs into DCIS, IDC, and benign. We evaluated the models on two independent test sets (n=1,382, n=548), achieving ROC areas under the curves (AUCs) up to 0.960 and 0.977 for DCIS and IDC, respectively.

Apparent Diffusion Coefficient-Based Convolutional Neural Network Model Can Be Better Than Sole Diffusion-Weighted Magnetic Resonance Imaging to Improve the Differentiation of Invasive Breast Cancer From Breast Ductal Carcinoma In Situ

Background and PurposeBreast ductal carcinoma in situ (DCIS) has no metastatic potential, and has better clinical outcomes compared with invasive breast cancer (IBC). Convolutional neural networks (CNNs) can adaptively extract features and may achieve higher efficiency in apparent diffusion coefficient (ADC)-based tumor invasion assessment. This study aimed to determine the feasibility of constructing an ADC-based CNN model to discriminate DCIS from IBC.MethodsThe study retrospectively enrolled 700 patients with primary breast cancer between March 2006 and June 2019 from our hospital, and randomly selected 560 patients as the training and validation sets (ratio of 3 to 1), and 140 patients as the internal test set. An independent external test set of 102 patients during July 2019 and May 2021 from a different scanner of our hospital was selected as the primary cohort using the same criteria. In each set, the status of tumor invasion was confirmed by pathologic examination. The CNN model was constructed to discriminate DCIS from IBC using the training and validation sets. The CNN model was evaluated using the internal and external tests, and compared with the discriminating performance using the mean ADC. The area under the curve (AUC), sensitivity, specificity, and accuracy were calculated to evaluate the performance of the previous model.ResultsThe AUCs of the ADC-based CNN model using the internal and external test sets were larger than those of the mean ADC (AUC: 0.977 vs. 0.866, P = 0.001; and 0.926 vs. 0.845, P = 0.096, respectively). Regarding the internal test set and external test set, the ADC-based CNN model yielded sensitivities of 0.893 and 0.873, specificities of 0.929 and 0.894, and accuracies of 0.907 and 0.902, respectively. Regarding the two test sets, the mean ADC showed sensitivities of 0.845 and 0.818, specificities of 0.821 and 0.829, and accuracies of 0.836 and 0.824, respectively. Using the ADC-based CNN model, the prediction only takes approximately one second for a single lesion.ConclusionThe ADC-based CNN model can improve the differentiation of IBC from DCIS with higher accuracy and less time.

Effect of Non-surgical Periodontal Therapy on Levels of Inflammatory Markers in Gingival Crevicular Fluid and Serum of Patients With Breast Cancer and Periodontitis.

Purpose: studies have demonstrated the positive impact of non-surgical periodontal therapy (NSPT) on the control of local and systemic infection/inflammation in normosystemic and systemically compromised patients, represented by the improvement of periodontal clinical parameters and reduction in the levels of inflammatory markers in the gingival crevicular fluid (GCF), saliva and serum. This study aimed to evaluate periodontal clinical parameters and inflammatory mediators in GCF and serum, before and after NSPT, in patients with periodontitis and breast cancer, before chemotherapy. Methods: seventeen women with histopathological diagnosis of invasive ductal carcinoma and periodontitis were submitted to the evaluation of clinical periodontal parameters (plaque index – PI, bleeding on probing – BOP, probing depth – PD, clinical attachment level – CAL) and submitted to scaling and root planing (SRP), at an interval of 24 hours. At the beginning of the study (baseline), before NSPT, samples of tumor microenvironment fluid (TM), GCF and peripheral blood (serum) were collected for the determination of inflammatory markers IL-1β, TNF-α, TGF-β and IL-17, using the LUMINEX methodology. Seven days after SRP, new GCF and serum samples were obtained and analyzed.Results: TGF-β levels were significantly decreased in GCF and serum (p<0.05), while IL-17 concentrations were statistically reduced in GCF (p<0.05). Conclusion: NSPT decreased local and systemic inflammatory markers and may be an important tool in the multidisciplinary approach of women with breast cancer and periodontitis before chemotherapy.

Ductal carcinoma in situ: a risk prediction model for the underestimation of invasive breast cancer

Patients with a biopsy diagnosis of ductal carcinoma in situ (DCIS) may be diagnosed with invasive breast cancer after excision. We evaluated the preoperative clinical and imaging predictors of DCIS that were associated with an upgrade to invasive carcinoma on final pathology and also compared the diagnostic performance of various statistical models. We reviewed the medical records; including mammography, ultrasound (US), and magnetic resonance imaging (MRI) findings; of 644 patients who were preoperatively diagnosed with DCIS and who underwent surgery between January 2012 and September 2018. Logistic regression and three machine learning methods were applied to predict DCIS underestimation. Among 644 DCIS biopsies, 161 (25%) underestimated invasive breast cancers. In multivariable analysis, suspicious axillary lymph nodes (LNs) on US (odds ratio [OR], 12.16; 95% confidence interval [CI], 4.94–29.95; P < 0.001) and high nuclear grade (OR, 1.90; 95% CI, 1.24–2.91; P = 0.003) were associated with underestimation. Cases with biopsy performed using vacuum-assisted biopsy (VAB) (OR, 0.42; 95% CI, 0.27–0.65; P < 0.001) and lesion size <2 cm on mammography (OR, 0.45; 95% CI, 0.22–0.90; P = 0.021) and MRI (OR, 0.29; 95% CI, 0.09–0.94; P = 0.037) were less likely to be upgraded. No significant differences in performance were observed between logistic regression and machine learning models. Our results suggest that biopsy device, high nuclear grade, presence of suspicious axillary LN on US, and lesion size on mammography or MRI were independent predictors of DCIS underestimation.

Diagnostic performance of superb microvascular imaging combined with shear-wave elastography in distinguishing invasive ductal carcinoma molecular subtypes

IntroductionTo explore the diagnostic value of combining superb microvascular imaging (SMI), shear-wave elastography (SWE), and Breast Imaging Reporting and Data System (BI-RADS) to distinguish different molecular subtypes of invasive ductal carcinoma (IDC).Material and methodsA total of 239 surgically confirmed IDC masses in 201 patients underwent conventional ultrasound, SMI, and SWE examination, the information such as echo pattern, posterior features, margins, SMI pixels, and hardness of the masses was recorded. According to the St. Gallen standard, breast masses were classified as Luminal A, Luminal B, HER2 overexpression, and triple-negative subtype. We further explored the differences between different molecular subtypes of IDC.ResultsLuminal A subtype had the following characteristics: low histologic grade, posterior acoustic shadowing (p= 0.019), spiculated margins (p<0.001) , and relatively soft. Luminal B subtype was characterized by low histological grade (p <0.0001), posterior acoustic shadowing or indifference, and indistinct margins. HER2 overexpression breast cancers were characterized by high histological grade, enhanced posterior acoustics or indifference, calcifications (p= 0.005), spiculated or indistinct margins, vascularity (p=0.005), and relative stiffness. Triple-negative breast cancers had the characteristics of high histological grade, posterior echogenic enhancement, lack of calcifications, circumscribed or microlobulated margins, low blood flow signals, and stiff tissue (p=0.013).ConclusionsOur study demonstrated the significant differences and trends among the IDC four subtypes by the combined application of SMI, SWE, and BI-RADS lexicon, which are of great significance for early diagnosis, selection of treatment methods, and evaluation of prognosis of IDC.

A Model to Predict Upstaging to Invasive Carcinoma in Patients Preoperatively Diagnosed with Low-Grade Ductal Carcinoma In Situ of the Breast

Background: We aimed to create a model of radiological and pathological criteria able to predict the upgrade rate of low-grade ductal carcinoma in situ (DCIS) to invasive carcinoma, in patients undergoing vacuum-assisted breast biopsy (VABB) and subsequent surgical excision. Methods: A total of 3100 VABBs were retrospectively reviewed, among which we reported 295 low-grade DCIS who subsequently underwent surgery. The association between patients’ features and the upgrade rate to invasive breast cancer (IBC) was evaluated by univariate and multivariate analysis. Finally, we developed a nomogram for predicting the upstage at surgery, according to the multivariate logistic regression model. Results: The overall upgrade rate to invasive carcinoma was 10.8%. At univariate analysis, the risk of upgrade was significantly lower in patients with greater age (p = 0.018), without post-biopsy residual lesion (p < 0.001), with a smaller post-biopsy residual lesion size (p < 0.001), and in the presence of low-grade DCIS only in specimens with microcalcifications (p = 0.002). According to the final multivariable model, the predicted probability of upstage at surgery was lower than 2% in 58 patients; among these 58 patients, only one (1.7%) upstage was observed, showing a good calibration of the model. Conclusions: An easy-to-use nomogram for predicting the upstage at surgery based on radiological and pathological criteria is able to identify patients with low-grade carcinoma in situ with low risk of upstaging to infiltrating carcinomas.

StackBC: Deep learning and transfer learning techniques based stacking approach for accurate Invasive Ductal Carcinoma classification using histology images

Artificial intelligence (AI) based computer-aided diagnostic (CAD) systems can effectively diagnose critical disease. AI-based detection of breast cancer (BC) through images data is more efficient and accurate than professional radiologists. However, the existing AI-based BC diagnosis methods have complexity in low prediction accuracy and high computation time. Due to these reasons, medical professionals are not employing the current proposed techniques in E-Healthcare to effectively diagnose the BC. To diagnose the breast cancer effectively need to incorporate advanced AI techniques based methods in diagnosis process. In this work, we proposed a deep learning based diagnosis method (StackBC) to detect breast cancer in the early stage for effective treatment and recovery. In particular, we have incorporated deep learning models including Convolutional neural network (CNN), Long short term memory (LSTM), and Gated recurrent unit (GRU) for the classification of Invasive Ductal Carcinoma (IDC). Additionally, data augmentation and transfer learning techniques have been incorporated for data set balancing and for effective training the model. To further improve the predictive performance of model we used stacking technique. Among the three base classifiers (CNN, LSTM, GRU) the predictive performance of GRU are better as compared to individual model. The GRU is selected as a meta classifier to distinguish between Non-IDC and IDC breast images. The method Hold-Out has been incorporated and the data set is split into 90% and 10% for training and testing of the model, respectively. Model evaluation metrics have been computed for model performance evaluation. To analyze the efficacy of the model, we have used breast histology images data set. Our experimental results demonstrated that the proposed StackBC method achieved improved performance by gaining 99.02% accuracy and 100% area under the receiver operating characteristics curve (AUC-ROC) compared to state-of-the-art methods. Due to the high performance of the proposed method, we recommend it for early recognition of breast cancer in E-Healthcare.