Classification of Mammographic Breast Microcalcifications Using a Deep Convolutional Neural Network: A BI-RADS-Based Approach

Deep learning performance for detection and classification of microcalcifications on mammography

BACKGROUND

Breast cancer screening through mammography is crucial for early detection, yet the demand for mammography services surpasses the capacity of radiologists. Artificial intelligence (AI) can assist in evaluating microcalcifications on mammography. We developed and tested an AI model for localizing and characterizing microcalcifications.

METHODS

Three expert radiologists annotated a dataset of mammograms using histology-based ground truth. The dataset was partitioned for training, validation, and testing. Three neural networks (AlexNet, ResNet18, and ResNet34) were trained and evaluated using specific metrics including receiver operating characteristics area under the curve (AUC), sensitivity, and specificity. The reported metrics were computed on the test set (10% of the whole dataset).

RESULTS

The dataset included 1,000 patients aged 21-73 years and 1,986 mammograms (180 density A, 220 density B, 380 density C, and 220 density D), with 389 malignant and 611 benign groups of microcalcifications. AlexNet achieved the best performance with 0.98 sensitivity, 0.89 specificity of, and 0.98 AUC for microcalcifications detection and 0.85 sensitivity, 0.89 specificity, and 0.94 AUC of for microcalcifications classification. For microcalcifications detection, ResNet18 and ResNet34 achieved 0.96 and 0.97 sensitivity, 0.91 and 0.90 specificity and 0.98 and 0.98 AUC, retrospectively. For microcalcifications classification, ResNet18 and ResNet34 exhibited 0.75 and 0.84 sensitivity, 0.85 and 0.84 specificity, and 0.88 and 0.92 AUC, respectively.

CONCLUSIONS

The developed AI models accurately detect and characterize microcalcifications on mammography.

RELEVANCE STATEMENT

AI-based systems have the potential to assist radiologists in interpreting microcalcifications on mammograms. The study highlights the importance of developing reliable deep learning models possibly applied to breast cancer screening.

KEY POINTS

• A novel AI tool was developed and tested to aid radiologists in the interpretation of mammography by accurately detecting and characterizing microcalcifications. • Three neural networks (AlexNet, ResNet18, and ResNet34) were trained, validated, and tested using an annotated dataset of 1,000 patients and 1,986 mammograms. • The AI tool demonstrated high accuracy in detecting/localizing and characterizing microcalcifications on mammography, highlighting the potential of AI-based systems to assist radiologists in the interpretation of mammograms.

The application of traditional machine learning and deep learning techniques in mammography: a review

Breast cancer, the most prevalent malignant tumor among women, poses a significant threat to patients' physical and mental well-being. Recent advances in early screening technology have facilitated the early detection of an increasing number of breast cancers, resulting in a substantial improvement in patients' overall survival rates. The primary techniques used for early breast cancer diagnosis include mammography, breast ultrasound, breast MRI, and pathological examination. However, the clinical interpretation and analysis of the images produced by these technologies often involve significant labor costs and rely heavily on the expertise of clinicians, leading to inherent deviations. Consequently, artificial intelligence(AI) has emerged as a valuable technology in breast cancer diagnosis. Artificial intelligence includes Machine Learning(ML) and Deep Learning(DL). By simulating human behavior to learn from and process data, ML and DL aid in lesion localization reduce misdiagnosis rates, and improve accuracy. This narrative review provides a comprehensive review of the current research status of mammography using traditional ML and DL algorithms. It particularly highlights the latest advancements in DL methods for mammogram image analysis and offers insights into future development directions.

Diagnostic accuracy of automated ACR BI-RADS breast density classification using deep convolutional neural networks

OBJECTIVES

High breast density is a well-known risk factor for breast cancer. This study aimed to develop and adapt two (MLO, CC) deep convolutional neural networks (DCNN) for automatic breast density classification on synthetic 2D tomosynthesis reconstructions.

METHODS

In total, 4605 synthetic 2D images (1665 patients, age: 57 ± 37 years) were labeled according to the ACR (American College of Radiology) density (A-D). Two DCNNs with 11 convolutional layers and 3 fully connected layers each, were trained with 70% of the data, whereas 20% was used for validation. The remaining 10% were used as a separate test dataset with 460 images (380 patients). All mammograms in the test dataset were read blinded by two radiologists (reader 1 with two and reader 2 with 11 years of dedicated mammographic experience in breast imaging), and the consensus was formed as the reference standard. The inter- and intra-reader reliabilities were assessed by calculating Cohen's kappa coefficients, and diagnostic accuracy measures of automated classification were evaluated.

RESULTS

The two models for MLO and CC projections had a mean sensitivity of 80.4% (95%-CI 72.2-86.9), a specificity of 89.3% (95%-CI 85.4-92.3), and an accuracy of 89.6% (95%-CI 88.1-90.9) in the differentiation between ACR A/B and ACR C/D. DCNN versus human and inter-reader agreement were both "substantial" (Cohen's kappa: 0.61 versus 0.63).

CONCLUSION

The DCNN allows accurate, standardized, and observer-independent classification of breast density based on the ACR BI-RADS system.

KEY POINTS

• A DCNN performs on par with human experts in breast density assessment for synthetic 2D tomosynthesis reconstructions. • The proposed technique may be useful for accurate, standardized, and observer-independent breast density evaluation of tomosynthesis.

DeepMiCa: Automatic segmentation and classification of breast MIcroCAlcifications from mammograms

BACKGROUND AND OBJECTIVE

Breast cancer is the world's most prevalent form of cancer. The survival rates have increased in the last years mainly due to factors such as screening programs for early detection, new insights on the disease mechanisms as well as personalised treatments. Microcalcifications are the only first detectable sign of breast cancer and diagnosis timing is strongly related to the chances of survival. Nevertheless microcalcifications detection and classification as benign or malignant lesions is still a challenging clinical task and their malignancy can only be proven after a biopsy procedure. We propose DeepMiCa, a fully automated and visually explainable deep-learning based pipeline for the analysis of raw mammograms with microcalcifications. Our aim is to propose a reliable decision support system able to guide the diagnosis and help the clinicians to better inspect borderline difficult cases.

METHODS

DeepMiCa is composed by three main steps: (1) Preprocessing of the raw scans (2) Automatic patch-based Semantic Segmentation using a UNet based network with a custom loss function appositely designed to deal with extremely small lesions (3) Classification of the detected lesions with a deep transfer-learning approach. Finally, state-of-the-art explainable AI methods are used to produce maps for a visual interpretation of the classification results. Each step of DeepMiCa is designed to address the main limitations of the previous proposed works resulting in a novel automated and accurate pipeline easily customisable to meet radiologists' needs.

RESULTS

The proposed segmentation and classification algorithms achieve an area under the ROC curve of 0.95 and 0.89 respectively. Compared to previously proposed works, this method does not require high performance computational resources and provides a visual explanation of the final classification results.

CONCLUSION

To conclude, we designed a novel fully automated pipeline for detection and classification of breast microcalcifications. We believe that the proposed system has the potential to provide a second opinion in the diagnosis process giving the clinicians the opportunity to quickly visualise and inspect relevant imaging characteristics. In the clinical practice the proposed decision support system could help reduce the rate of misclassified lesions and consequently the number of unnecessary biopsies.

Automatic and standardized quality assurance of digital mammography and tomosynthesis with deep convolutional neural networks

OBJECTIVES

The aim of this study was to develop and validate a commercially available AI platform for the automatic determination of image quality in mammography and tomosynthesis considering a standardized set of features.

MATERIALS AND METHODS

In this retrospective study, 11,733 mammograms and synthetic 2D reconstructions from tomosynthesis of 4200 patients from two institutions were analyzed by assessing the presence of seven features which impact image quality in regard to breast positioning. Deep learning was applied to train five dCNN models on features detecting the presence of anatomical landmarks and three dCNN models for localization features. The validity of models was assessed by the calculation of the mean squared error in a test dataset and was compared to the reading by experienced radiologists.

RESULTS

Accuracies of the dCNN models ranged between 93.0% for the nipple visualization and 98.5% for the depiction of the pectoralis muscle in the CC view. Calculations based on regression models allow for precise measurements of distances and angles of breast positioning on mammograms and synthetic 2D reconstructions from tomosynthesis. All models showed almost perfect agreement compared to human reading with Cohen's kappa scores above 0.9.

CONCLUSIONS

An AI-based quality assessment system using a dCNN allows for precise, consistent and observer-independent rating of digital mammography and synthetic 2D reconstructions from tomosynthesis. Automation and standardization of quality assessment enable real-time feedback to technicians and radiologists that shall reduce a number of inadequate examinations according to PGMI (Perfect, Good, Moderate, Inadequate) criteria, reduce a number of recalls and provide a dependable training platform for inexperienced technicians.

Validation of Combined Deep Learning Triaging and Computer-Aided Diagnosis in 2901 Breast MRI Examinations From the Second Screening Round of the Dense Tissue and Early Breast Neoplasm Screening Trial

OBJECTIVES

Computer-aided triaging (CAT) and computer-aided diagnosis (CAD) of screening breast magnetic resonance imaging have shown potential to reduce the workload of radiologists in the context of dismissing normal breast scans and dismissing benign disease in women with extremely dense breasts. The aim of this study was to validate the potential of integrating CAT and CAD to reduce workload and workup on benign lesions in the second screening round of the DENSE trial, without missing cancer.

METHODS

We included 2901 breast magnetic resonance imaging scans, obtained from 8 hospitals in the Netherlands. Computer-aided triaging and CAD were previously developed on data from the first screening round. Computer-aided triaging dismissed examinations without lesions. Magnetic resonance imaging examinations triaged to radiological reading were counted and subsequently processed by CAD. The number of benign lesions correctly classified by CAD was recorded. The false-positive fraction of the CAD was compared with that of unassisted radiological reading in the second screening round. Receiver operating characteristics (ROC) analysis was performed and the generalizability of CAT and CAD was assessed by comparing results from first and second screening rounds.

RESULTS

Computer-aided triaging dismissed 950 of 2901 (32.7%) examinations with 49 lesions in total; none were malignant. Subsequent CAD classified 132 of 285 (46.3%) lesions as benign without misclassifying any malignant lesion. Together, CAT and CAD yielded significantly fewer false-positive lesions, 53 of 109 (48.6%) and 89 of 109 (78.9%), respectively ( P = 0.001), than radiological reading alone. Computer-aided triaging had a smaller area under the ROC curve in the second screening round compared with the first, 0.83 versus 0.76 ( P = 0.001), but this did not affect the negative predictive value at the 100% sensitivity operating threshold. Computer-aided diagnosis was not associated with significant differences in area under the ROC curve (0.857 vs 0.753, P = 0.08). At the operating thresholds, the specificities of CAT (39.7% vs 41.0%, P = 0.70) and CAD (41.0% vs 38.2%, P = 0.62) were successfully reproduced in the second round.

CONCLUSION

The combined application of CAT and CAD showed potential to reduce workload of radiologists and to reduce number of biopsies on benign lesions. Computer-aided triaging (CAT) correctly dismissed 950 of 2901 (32.7%) examinations with 49 lesions in total; none were malignant. Subsequent computer-aided diagnosis (CAD) classified 132 of 285 (46.3%) lesions as benign without misclassifying any malignant lesion. Together, CAT and CAD yielded significantly fewer false-positive lesions, 53 of 109 (48.6%) and 89 of 109 (78.9%), respectively ( P = 0.001), than radiological reading alone.

Detection of microcalcifications in photon-counting dedicated breast-CT using a deep convolutional neural network: Proof of principle

OBJECTIVE

In this study, we investigate the feasibility of a deep Convolutional Neural Network (dCNN), trained with mammographic images, to detect and classify microcalcifications (MC) in breast-CT (BCT) images.

METHODS

This retrospective single-center study was approved by the local ethics committee. 3518 icons generated from 319 mammograms were classified into three classes: "no MC" (1121), "probably benign MC" (1332), and "suspicious MC" (1065). A dCNN was trained (70% of data), validated (20%), and tested on a "real-world" dataset (10%). The diagnostic performance of the dCNN was tested on a subset of 60 icons, generated from 30 mammograms and 30 breast-CT images, and compared to human reading. ROC analysis was used to calculate diagnostic performance. Moreover, colored probability maps for representative BCT images were calculated using a sliding-window approach.

RESULTS

The dCNN reached an accuracy of 98.8% on the "real-world" dataset. The accuracy on the subset of 60 icons was 100% for mammographic images, 60% for "no MC", 80% for "probably benign MC" and 100% for "suspicious MC". Intra-class correlation between the dCNN and the readers was almost perfect (0.85). Kappa values between the two readers (0.93) and the dCNN were almost perfect (reader 1: 0.85 and reader 2: 0.82). The sliding-window approach successfully detected suspicious MC with high image quality. The diagnostic performance of the dCNN to classify benign and suspicious MC was excellent with an AUC of 93.8% (95% CI 87, 4%-100%).

CONCLUSION

Deep convolutional networks can be used to detect and classify benign and suspicious MC in breast-CT images.

Artificial Intelligence for Indication of Invasive Assessment of Calcifications in Mammography Screening

PURPOSE

Lesion-related evaluation of the diagnostic performance of an individual artificial intelligence (AI) system to assess mamographically detected and histologically proven calcifications.

MATERIALS AND METHODS

This retrospective study included 634 women of one screening unit (July 2012 - June 2018) who completed the invasive assessment of calcifications. For each leasion, the AI-system calculated a score between 0 and 98. Lesions scored > 0 were classified as AI-positive. The performance of the system was evaluated based on its positive predictive value of invasive assessment (PPV3), the false-negative rate and the true-negative rate.

RESULTS

The PPV3 increased across the categories (readers: 4a: 21.2 %, 4b: 57.7 %, 5: 100 %, overall 30.3 %; AI: 4a: 20.8 %, 4b: 57.8 %, 5: 100 %, overall: 30.7 %). The AI system yielded a false-negative rate of 7.2 % (95 %-CI: 4.3 %: 11.4 %) and a true-negative rate of 9.1 % (95 %-CI: 6.6 %; 11.9 %). These rates were highest in category 4a, 12.5 % and 10.4 % retrospectively. The lowest median AI score was observed for benign lesions (61, interquartile range (IQR): 45-74). Invasive cancers yielded the highest median AI score (81, IQR: 64-86). Median AI scores for ductal carcinoma in situ were: 74 (IQR: 63-84) for low grade, 70 (IQR: 52-79) for intermediate grade and 74 (IQR: 66-83) for high grade.

CONCLUSION

At the lowest threshold, the AI system yielded calcification-related PPV3 values that increased across categories, similar as seen in human evaluation. The strongest loss in AI-based breast cancer detection was observed for invasively assessed calcifications with the lowest suspicion of malignancy, yet with a comparable decrease in the false-positive rate. An AI-score based stratification of malignant lesions could not be determined.

KEY POINTS

· The AI-based PPV3 for calcifications is comparable to human assessment.. · AI showed a lower detection performance of screen-positive and screen-negative lesions in category 4a.. · Histological subgroups could not be discriminated by AI scores..

CITATION FORMAT

· Weigel S, Brehl AK, Heindel W et al. Artificial Intelligence for Indication of Invasive Assessment of Calcifications in Mammography Screening. Fortschr Röntgenstr 2023; 195: 38 - 46.

Fully automatic classification of automated breast ultrasound (ABUS) imaging according to BI-RADS using a deep convolutional neural network

PURPOSE

The aim of this study was to develop and test a post-processing technique for detection and classification of lesions according to the BI-RADS atlas in automated breast ultrasound (ABUS) based on deep convolutional neural networks (dCNNs).

METHODS AND MATERIALS

In this retrospective study, 645 ABUS datasets from 113 patients were included; 55 patients had lesions classified as high malignancy probability. Lesions were categorized in BI-RADS 2 (no suspicion of malignancy), BI-RADS 3 (probability of malignancy < 3%), and BI-RADS 4/5 (probability of malignancy > 3%). A deep convolutional neural network was trained after data augmentation with images of lesions and normal breast tissue, and a sliding-window approach for lesion detection was implemented. The algorithm was applied to a test dataset containing 128 images and performance was compared with readings of 2 experienced radiologists.

RESULTS

Results of calculations performed on single images showed accuracy of 79.7% and AUC of 0.91 [95% CI: 0.85-0.96] in categorization according to BI-RADS. Moderate agreement between dCNN and ground truth has been achieved (κ: 0.57 [95% CI: 0.50-0.64]) what is comparable with human readers. Analysis of whole dataset improved categorization accuracy to 90.9% and AUC of 0.91 [95% CI: 0.77-1.00], while achieving almost perfect agreement with ground truth (κ: 0.82 [95% CI: 0.69-0.95]), performing on par with human readers. Furthermore, the object localization technique allowed the detection of lesion position slice-wise.

CONCLUSIONS

Our results show that a dCNN can be trained to detect and distinguish lesions in ABUS according to the BI-RADS classification with similar accuracy as experienced radiologists.

KEY POINTS

• A deep convolutional neural network (dCNN) was trained for classification of ABUS lesions according to the BI-RADS atlas. • A sliding-window approach allows accurate automatic detection and classification of lesions in ABUS examinations.

BI-RADS-Based Classification of Mammographic Soft Tissue Opacities Using a Deep Convolutional Neural Network

The aim of this study was to investigate the potential of a machine learning algorithm to classify breast cancer solely by the presence of soft tissue opacities in mammograms, independent of other morphological features, using a deep convolutional neural network (dCNN). Soft tissue opacities were classified based on their radiological appearance using the ACR BI-RADS atlas. We included 1744 mammograms from 438 patients to create 7242 icons by manual labeling. The icons were sorted into three categories: “no opacities” (BI-RADS 1), “probably benign opacities” (BI-RADS 2/3) and “suspicious opacities” (BI-RADS 4/5). A dCNN was trained (70% of data), validated (20%) and finally tested (10%). A sliding window approach was applied to create colored probability maps for visual impression. Diagnostic performance of the dCNN was compared to human readout by experienced radiologists on a “real-world” dataset. The accuracies of the models on the test dataset ranged between 73.8% and 89.8%. Compared to human readout, our dCNN achieved a higher specificity (100%, 95% CI: 85.4–100%; reader 1: 86.2%, 95% CI: 67.4–95.5%; reader 2: 79.3%, 95% CI: 59.7–91.3%), and the sensitivity (84.0%, 95% CI: 63.9–95.5%) was lower than that of human readers (reader 1:88.0%, 95% CI: 67.4–95.4%; reader 2:88.0%, 95% CI: 67.7–96.8%). In conclusion, a dCNN can be used for the automatic detection as well as the standardized and observer-independent classification of soft tissue opacities in mammograms independent of the presence of microcalcifications. Human decision making in accordance with the BI-RADS classification can be mimicked by artificial intelligence.

Detecting Abnormal Axillary Lymph Nodes on Mammograms Using a Deep Convolutional Neural Network

The purpose of this study was to determine the feasibility of a deep convolutional neural network (dCNN) to accurately detect abnormal axillary lymph nodes on mammograms. In this retrospective study, 107 mammographic images in mediolateral oblique projection from 74 patients were labeled to three classes: (1) “breast tissue”, (2) “benign lymph nodes”, and (3) “suspicious lymph nodes”. Following data preprocessing, a dCNN model was trained and validated with 5385 images. Subsequently, the trained dCNN was tested on a “real-world” dataset and the performance compared to human readers. For visualization, colored probability maps of the classification were calculated using a sliding window approach. The accuracy was 98% for the training and 99% for the validation set. Confusion matrices of the “real-world” dataset for the three classes with radiological reports as ground truth yielded an accuracy of 98.51% for breast tissue, 98.63% for benign lymph nodes, and 95.96% for suspicious lymph nodes. Intraclass correlation of the dCNN and the readers was excellent (0.98), and Kappa values were nearly perfect (0.93–0.97). The colormaps successfully detected abnormal lymph nodes with excellent image quality. In this proof-of-principle study in a small patient cohort from a single institution, we found that deep convolutional networks can be trained with high accuracy and reliability to detect abnormal axillary lymph nodes on mammograms.

The effect of the use of the Gail model on breast cancer diagnosis in BIRADs 4a cases

Objective: The BI-RADS classification system and the Gail Model are the scoring systems that contribute to the diagnosis of breast cancer. The aim of the study was to determine the contribution of Gail Model to the diagnosis of breast lesions that were radiologically categorized as BI-RADS 4A.

Material and Methods: We retrospectively examined the medical records of 320 patients between January 2011 and December 2020 whose lesions had been categorized as BI-RADS 4A. Radiological parameters of breast lesions and clinical parameters according to the Gail Model were collected. The relationship between malignant BI-RADS 4A lesions and radiological and clinical parameters was evaluated. In addition, the effect of the Gail Model on diagnosis in malignant BI-RADS 4A lesions was evaluated.

Results: Among radiological features, there were significant differences between lesion size, contour, microcalcification content, echogenicity, and presence of ectasia with respect to the pathological diagnosis (p< 0.05). No significant difference was found between the lesions’ pathological diagnosis and the patients’ Gail score (p> 0.05). An analysis of the features of the Gail model revealed that there was no significant difference between the age of menarche, age at first live birth, presence of a first-degree relative with breast cancer, and a history of breast biopsy and the pathological diagnosis (p> 0.05).

Conclusion: As a conclusion Gail Model does not contribute to the diagnosis of BC, especially in patients with BI-RADS 4A lesions.

Diagnostic Performance of Artificial Intelligence-Based Computer-Aided Diagnosis for Breast Microcalcification on Mammography

The present study evaluated the diagnostic performance of artificial intelligence-based computer-aided diagnosis (AI-CAD) compared to that of dedicated breast radiologists in characterizing suspicious microcalcification on mammography. We retrospectively analyzed 435 unilateral mammographies from 420 patients (286 benign; 149 malignant) undergoing biopsy for suspicious microcalcification from June 2003 to November 2019. Commercial AI-CAD was applied to the mammography images, and malignancy scores were calculated. Diagnostic performance was compared between radiologists and AI-CAD using the area under the receiving operator characteristics curve (AUC). The AUCs of radiologists and AI-CAD were not significantly different (0.722 vs. 0.745, p = 0.393). The AUCs of the adjusted category were 0.726, 0.744, and 0.756 with cutoffs of 2%, 10%, and 38.03% for AI-CAD, respectively, which were all significantly higher than those for radiologists alone (all p < 0.05). None of the 27 cases downgraded to category 3 with a cutoff of 2% were confirmed as malignant on pathological analysis, suggesting that unnecessary biopsies could be avoided. Our findings suggest that the diagnostic performance of AI-CAD in characterizing suspicious microcalcification on mammography was similar to that of the radiologists, indicating that it may aid in making clinical decisions regarding the treatment of breast microcalcification.

Machine learning predictive model for severe COVID-19

To develop a modified predictive model for severe COVID-19 in people infected with Sars-Cov-2. We developed the predictive model for severe patients of COVID-19 based on the clinical date from the Tumor Center of Union Hospital affiliated with Tongji Medical College, China. A total of 151 cases from Jan. 26 to Mar. 20, 2020, were included. Then we followed 5 steps to predict and evaluate the model: data preprocessing, data splitting, feature selection, model building, prevention of overfitting, and Evaluation, and combined with artificial neural network algorithms. We processed the results in the 5 steps. In feature selection, ALB showed a strong negative correlation (r = 0.771, P < 0.001) whereas GLB (r = 0.661, P < 0.001) and BUN (r = 0.714, P < 0.001) showed a strong positive correlation with severity of COVID-19. TensorFlow was subsequently applied to develop a neural network model. The model achieved good prediction performance, with an area under the curve value of 0.953(0.889-0.982). Our results showed its outstanding performance in prediction. GLB and BUN may be two risk factors for severe COVID-19. Our findings could be of great benefit in the future treatment of patients with COVID-19 and will help to improve the quality of care in the long term. This model has great significance to rationalize early clinical interventions and improve the cure rate.