Digital reference object toolkit of breast DCE MRI for quantitative evaluation of image reconstruction and analysis methods.

PURPOSE: To develop a digital reference object (DRO) toolkit to generate realistic breast DCE-MRI data for quantitative assessment of image reconstruction and data analysis methods. METHODS: A simulation framework in a form of DRO toolkit has been developed using the ultrafast and conventional breast DCE-MRI data of 53 women with malignant (n = 25) or benign (n = 28) lesions. We segmented five anatomical regions and performed pharmacokinetic analysis to determine the ranges of pharmacokinetic parameters for each segmented region. A database of the segmentations and their pharmacokinetic parameters is included in the DRO toolkit that can generate a large number of realistic breast DCE-MRI data. We provide two potential examples for our DRO toolkit: assessing the accuracy of an image reconstruction method using undersampled simulated radial k-space data and assessing the impact of the B 1 + $$ {\mathrm{B}}_1^{+} $$ field inhomogeneity on estimated parameters. RESULTS: The estimated pharmacokinetic parameters for each region showed agreement with previously reported values. For the assessment of the reconstruction method, it was found that the temporal regularization resulted in significant underestimation of estimated parameters by up to 57% and 10% with the weighting factor λ = 0.1 and 0.01, respectively. We also demonstrated that spatial discrepancy of v p $$ {v}_p $$ and PS $$ \mathrm{PS} $$ increase to about 33% and 51% without correction for B 1 + $$ {\mathrm{B}}_1^{+} $$ field. CONCLUSION: We have developed a DRO toolkit that includes realistic morphology of tumor lesions along with the expected pharmacokinetic parameter ranges. This simulation framework can generate many images for quantitative assessment of DCE-MRI reconstruction and analysis methods.

An Efficient Deep Neural Network to Classify Large 3D Images With Small Objects.

3D imaging enables accurate diagnosis by providing spatial information about organ anatomy. However, using 3D images to train AI models is computationally challenging because they consist of 10x or 100x more pixels than their 2D counterparts. To be trained with high-resolution 3D images, convolutional neural networks resort to downsampling them or projecting them to 2D. We propose an effective alternative, a neural network that enables efficient classification of full-resolution 3D medical images. Compared to off-the-shelf convolutional neural networks, our network, 3D Globally-Aware Multiple Instance Classifier (3D-GMIC), uses 77.98%-90.05% less GPU memory and 91.23%-96.02% less computation. While it is trained only with image-level labels, without segmentation labels, it explains its predictions by providing pixel-level saliency maps. On a dataset collected at NYU Langone Health, including 85,526 patients with full-field 2D mammography (FFDM), synthetic 2D mammography, and 3D mammography, 3D-GMIC achieves an AUC of 0.831 (95% CI: 0.769-0.887) in classifying breasts with malignant findings using 3D mammography. This is comparable to the performance of GMIC on FFDM (0.816, 95% CI: 0.737-0.878) and synthetic 2D (0.826, 95% CI: 0.754-0.884), which demonstrates that 3D-GMIC successfully classified large 3D images despite focusing computation on a smaller percentage of its input compared to GMIC. Therefore, 3D-GMIC identifies and utilizes extremely small regions of interest from 3D images consisting of hundreds of millions of pixels, dramatically reducing associated computational challenges. 3D-GMIC generalizes well to BCS-DBT, an external dataset from Duke University Hospital, achieving an AUC of 0.848 (95% CI: 0.798-0.896).

Problem-solving Breast MRI.

Breast MRI has high sensitivity and negative predictive value, making it well suited to problem solving when other imaging modalities or physical examinations yield results that are inconclusive for the presence of breast cancer. Indications for problem-solving MRI include equivocal or uncertain imaging findings at mammography and/or US; suspicious nipple discharge or skin changes suspected to represent an abnormality when conventional imaging results are negative for cancer; lesions categorized as Breast Imaging Reporting and Data System 4, which are not amenable to biopsy; and discordant radiologic-pathologic findings after biopsy. MRI should not precede or replace careful diagnostic workup with mammography and US and should not be used when a biopsy can be safely performed. The role of MRI in characterizing calcifications is controversial, and management of calcifications should depend on their mammographic appearance because ductal carcinoma in situ may not appear enhancing on MR images. In addition, ductal carcinoma in situ detected solely with MRI is not associated with a higher likelihood of an upgrade to invasive cancer compared with ductal carcinoma in situ detected with other modalities. MRI for triage of high-risk lesions is a subject of ongoing investigation, with a possible future role for MRI in decreasing excisional biopsies. The accuracy of MRI is likely to increase with the use of advanced techniques such as deep learning, which will likely expand the indications for problem-solving MRI. ©RSNA, 2023 Quiz questions for this article are available in the supplemental material.

Improving Information Extraction from Pathology Reports using Named Entity Recognition.

Pathology reports are considered the gold standard in medical research due to their comprehensive and accurate diagnostic information. Natural language processing (NLP) techniques have been developed to automate information extraction from pathology reports. However, existing studies suffer from two significant limitations. First, they typically frame their tasks as report classification, which restricts the granularity of extracted information. Second, they often fail to generalize to unseen reports due to variations in language, negation, and human error. To overcome these challenges, we propose a BERT (bidirectional encoder representations from transformers) named entity recognition (NER) system to extract key diagnostic elements from pathology reports. We also introduce four data augmentation methods to improve the robustness of our model. Trained and evaluated on 1438 annotated breast pathology reports, acquired from a large medical center in the United States, our BERT model trained with data augmentation achieves an entity F1-score of 0.916 on an internal test set, surpassing the BERT baseline (0.843). We further assessed the model's generalizability using an external validation dataset from the United Arab Emirates, where our model maintained satisfactory performance (F1-score 0.860). Our findings demonstrate that our NER systems can effectively extract fine-grained information from widely diverse medical reports, offering the potential for large-scale information extraction in a wide range of medical and AI research. We publish our code at https://github.com/nyukat/pathology_extraction.

PACS-integrated machine learning breast density classifier: clinical validation.

OBJECTIVE: To test the performance of a novel machine learning-based breast density tool. The tool utilizes a convolutional neural network to predict the BI-RADS based density assessment of a study. The clinical density assessments of 33,000 mammographic examinations (164,000 images) from one academic medical center (Site A) were used for training. MATERIALS AND METHODS: This was an IRB approved HIPAA compliant study performed at two academic medical centers. The validation data set was composed of 500 studies from one site (Site A) and 700 from another (Site B). At Site A, each study was assessed by three breast radiologists and the majority (consensus) assessment was used as truth. At Site B, if the tool agreed with the clinical reading, then it was considered to have correctly predicted the clinical reading. In cases where the tool and the clinical reading disagreed, then the study was evaluated by three radiologists and the consensus reading was used as the clinical reading. RESULTS: For the classification into the four categories of the Breast Imaging Reporting and Data System (BI-RADS®), the AI classifier had an accuracy of 84.6% at Site A and 89.7% at Site B. For binary classification (dense vs. non-dense), the AI classifier had an accuracy of 94.4% at Site A and 97.4% at Site B. In no case did the classifier disagree with the consensus reading by more than one category. CONCLUSIONS: The automated breast density tool showed high agreement with radiologists' assessments of breast density.

Women 75 Years Old or Older: To Screen or Not to Screen?

Breast cancer is the most common cancer in women, with the incidence rising substantially with age. Older women are a vulnerable population at increased risk of developing and dying from breast cancer. However, women aged 75 years and older were excluded from all randomized controlled screening trials, so the best available data regarding screening benefits and risks in this age group are from observational studies and modeling predictions. Benefits of screening in older women are the same as those in younger women: early detection of smaller lower-stage cancers, resulting in less invasive treatment and lower morbidity and mortality. Mammography performs significantly better in older women with higher sensitivity, specificity, cancer detection rate, and positive predictive values, accompanied by lower recall rates and false positives. The overdiagnosis rate is low, with benefits outweighing risks until age 90 years. Although there are conflicting national and international guidelines about whether to continue screening mammography in women beyond age 74 years, clinicians can use shared decision making to help women make decisions about screening and fully engage them in the screening process. For women aged 75 years and older in good health, continuing annual screening mammography will save the most lives. An informed discussion of the benefits and risks of screening mammography in older women needs to include each woman's individual values, overall health status, and comorbidities. This article will review the benefits, risks, and controversies surrounding screening mammography in women 75 years old and older and compare the current recommendations for screening this population from national and international professional organizations. ©RSNA, 2023 Quiz questions for this article are available through the Online Learning Center.

New Horizons: Artificial Intelligence for Digital Breast Tomosynthesis.

The use of digital breast tomosynthesis (DBT) in breast cancer screening has become widely accepted, facilitating increased cancer detection and lower recall rates compared with those achieved by using full-field digital mammography (DM). However, the use of DBT, as compared with DM, raises new challenges, including a larger number of acquired images and thus longer interpretation times. While most current artificial intelligence (AI) applications are developed for DM, there are multiple potential opportunities for AI to augment the benefits of DBT. During the diagnostic steps of lesion detection, characterization, and classification, AI algorithms may not only assist in the detection of indeterminate or suspicious findings but also aid in predicting the likelihood of malignancy for a particular lesion. During image acquisition and processing, AI algorithms may help reduce radiation dose and improve lesion conspicuity on synthetic two-dimensional DM images. The use of AI algorithms may also improve workflow efficiency and decrease the radiologist's interpretation time. There has been significant growth in research that applies AI to DBT, with several algorithms approved by the U.S. Food and Drug Administration for clinical implementation. Further development of AI models for DBT has the potential to lead to improved practice efficiency and ultimately improved patient health outcomes of breast cancer screening and diagnostic evaluation. See the invited commentary by Bahl in this issue. ©RSNA, 2022.

Improving breast cancer diagnostics with deep learning for MRI.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has a high sensitivity in detecting breast cancer but often leads to unnecessary biopsies and patient workup. We used a deep learning (DL) system to improve the overall accuracy of breast cancer diagnosis and personalize management of patients undergoing DCE-MRI. On the internal test set (n = 3936 exams), our system achieved an area under the receiver operating characteristic curve (AUROC) of 0.92 (95% CI: 0.92 to 0.93). In a retrospective reader study, there was no statistically significant difference (P = 0.19) between five board-certified breast radiologists and the DL system (mean ΔAUROC, +0.04 in favor of the DL system). Radiologists' performance improved when their predictions were averaged with DL's predictions [mean ΔAUPRC (area under the precision-recall curve), +0.07]. We demonstrated the generalizability of the DL system using multiple datasets from Poland and the United States. An additional reader study on a Polish dataset showed that the DL system was as robust to distribution shift as radiologists. In subgroup analysis, we observed consistent results across different cancer subtypes and patient demographics. Using decision curve analysis, we showed that the DL system can reduce unnecessary biopsies in the range of clinically relevant risk thresholds. This would lead to avoiding biopsies yielding benign results in up to 20% of all patients with BI-RADS category 4 lesions. Last, we performed an error analysis, investigating situations where DL predictions were mostly incorrect. This exploratory work creates a foundation for deployment and prospective analysis of DL-based models for breast MRI.

Advances in Abbreviated Breast MRI and Ultrafast Imaging.

Abbreviated breast MRI is an emerging technique that is being incorporated into clinical practice for breast cancer imaging and screening. Conventional breast MRI includes barriers such as high examination cost and lengthy examination times which make its use in the screening setting challenging. Abbreviated MRI aims to address these pitfalls by reducing overall examination time and increasing accessibility to MRI while preserving diagnostic accuracy. Sequences selected for abbreviated MRI protocols allow for preserved accuracy in breast cancer detection and characterization. Novel techniques such as ultrafast imaging are being used to provide kinetic information from early post-contrast imaging.

Differences between human and machine perception in medical diagnosis.

Deep neural networks (DNNs) show promise in image-based medical diagnosis, but cannot be fully trusted since they can fail for reasons unrelated to underlying pathology. Humans are less likely to make such superficial mistakes, since they use features that are grounded on medical science. It is therefore important to know whether DNNs use different features than humans. Towards this end, we propose a framework for comparing human and machine perception in medical diagnosis. We frame the comparison in terms of perturbation robustness, and mitigate Simpson's paradox by performing a subgroup analysis. The framework is demonstrated with a case study in breast cancer screening, where we separately analyze microcalcifications and soft tissue lesions. While it is inconclusive whether humans and DNNs use different features to detect microcalcifications, we find that for soft tissue lesions, DNNs rely on high frequency components ignored by radiologists. Moreover, these features are located outside of the region of the images found most suspicious by radiologists. This difference between humans and machines was only visible through subgroup analysis, which highlights the importance of incorporating medical domain knowledge into the comparison.

Estimation of the capillary level input function for dynamic contrast-enhanced MRI of the breast using a deep learning approach.

PURPOSE: To develop a deep learning approach to estimate the local capillary-level input function (CIF) for pharmacokinetic model analysis of DCE-MRI. METHODS: A deep convolutional network was trained with numerically simulated data to estimate the CIF. The trained network was tested using simulated lesion data and used to estimate voxel-wise CIF for pharmacokinetic model analysis of breast DCE-MRI data using an abbreviated protocol from women with malignant (n = 25) and benign (n = 28) lesions. The estimated parameters were used to build a logistic regression model to detect the malignancy. RESULT: The pharmacokinetic parameters estimated using the network-predicted CIF from our breast DCE data showed significant differences between the malignant and benign groups for all parameters. Testing the diagnostic performance with the estimated parameters, the conventional approach with arterial input function (AIF) showed an area under the curve (AUC) between 0.76 and 0.87, and the proposed approach with CIF demonstrated similar performance with an AUC between 0.79 and 0.81. CONCLUSION: This study shows the feasibility of estimating voxel-wise CIF using a deep neural network. The proposed approach could eliminate the need to measure AIF manually without compromising the diagnostic performance to detect the malignancy in the clinical setting.

Reducing False-Positive Biopsies using Deep Neural Networks that Utilize both Local and Global Image Context of Screening Mammograms.

Breast cancer is the most common cancer in women, and hundreds of thousands of unnecessary biopsies are done around the world at a tremendous cost. It is crucial to reduce the rate of biopsies that turn out to be benign tissue. In this study, we build deep neural networks (DNNs) to classify biopsied lesions as being either malignant or benign, with the goal of using these networks as second readers serving radiologists to further reduce the number of false-positive findings. We enhance the performance of DNNs that are trained to learn from small image patches by integrating global context provided in the form of saliency maps learned from the entire image into their reasoning, similar to how radiologists consider global context when evaluating areas of interest. Our experiments are conducted on a dataset of 229,426 screening mammography examinations from 141,473 patients. We achieve an AUC of 0.8 on a test set consisting of 464 benign and 136 malignant lesions.

Diffusion weighted imaging for evaluation of breast lesions: Comparison between high b-value single-shot and routine readout-segmented sequences at 3 T.

PURPOSE: In this study, we compare readout-segmented echo-planar imaging (rs-EPI) Diffusion Weighted Imaging (DWI) to a work-in-progress single-shot EPI with modified Inversion Recovery Background Suppression (ss-EPI-mIRBS) sequence at 3 T using a b-value of 2000 s/mm2 on image quality, lesion visibility and evaluation time. METHOD: From September 2017 to December 2018, 23 women (one case used for training) with known breast cancer were included in this study, after providing signed informed consent. Women were scanned with the conventional rs-EPI sequence and the work-in-progress ss-EPI-mIRBS during the same examination. Four breast radiologists (4-13 years of experience) independently scored both series for overall image quality (1: extremely poor to 9: excellent). All lesions (47 in total, 36 malignant, and 11 benign and high-risk) were evaluated for visibility (1: not visible, 2: visible if location is given, 3: visible) and probability of malignancy (BI-RADS 1 to 5). ADC values were determined by measuring signal intensity in the lesions using dynamic contrast-enhanced (DCE) images for reference. Evaluation times for all assessments were automatically recorded. Results were analyzed using the visual grading characteristics (VGC) and the resulting area under the curve (AUCVGC) method. Statistical analysis was performed in SPSS, with McNemar tests, and paired t-tests used for comparison. RESULTS: No significant differences were detected between the two sequences in image quality (AUCVGC: 0.398, p = 0.087) and lesion visibility (AUCVGC: 0.534, p = 0.336) scores. Lesion characteristics (e.g benign and high-risk, versus malignant; small (≤10 mm) vs. larger (>10 mm)) did not result in different image quality or lesion visibility between sequences. Sensitivity (rs-EPI: 72.2% vs. ss-EPImIRBS: 78.5%, p = 0.108) and specificity (70.5% vs. 56.8%, p = 0.210, respectively) were comparable. In both sequences the mean ADC value was higher for benign and high-risk lesions than for malignant lesions (ss-EPI-mIRBS: p = 0.022 and rs-EPI: p = 0.055). On average, ss-EPI-mIRBS resulted in decreased overall reading time by 7.7 s/case (p = 0.067); a reduction of 17%. For malignant lesions, average reading time was significantly shorter using ss-EPI-mIRBS compared to rs-EPI (64.0 s/lesion vs. 75.9 s/lesion, respectively, p = 0.039). CONCLUSION: Based on this study, the ss-EPI sequence using a b-value of 2000 s/mm2 enables for a mIRBS acquisition with quality and lesion conspicuity that is comparable to conventional rs-EPI, but with a decreased reading time.

Artificial intelligence system reduces false-positive findings in the interpretation of breast ultrasound exams.

Though consistently shown to detect mammographically occult cancers, breast ultrasound has been noted to have high false-positive rates. In this work, we present an AI system that achieves radiologist-level accuracy in identifying breast cancer in ultrasound images. Developed on 288,767 exams, consisting of 5,442,907 B-mode and Color Doppler images, the AI achieves an area under the receiver operating characteristic curve (AUROC) of 0.976 on a test set consisting of 44,755 exams. In a retrospective reader study, the AI achieves a higher AUROC than the average of ten board-certified breast radiologists (AUROC: 0.962 AI, 0.924 ± 0.02 radiologists). With the help of the AI, radiologists decrease their false positive rates by 37.3% and reduce requested biopsies by 27.8%, while maintaining the same level of sensitivity. This highlights the potential of AI in improving the accuracy, consistency, and efficiency of breast ultrasound diagnosis.

Breast MRI for Evaluation of Response to Neoadjuvant Therapy.

Neoadjuvant therapy is increasingly being used to treat early-stage triple-negative and human epidermal growth factor 2-overexpressing breast cancers, as well as locally advanced and inflammatory breast cancers. The rationales for neoadjuvant therapy are to shrink tumor size and potentially decrease the extent of surgery, to serve as an in vivo test of response to therapy, and to reveal prognostic information for the patient. MRI is the most accurate modality to demonstrate response to therapy and to help ensure accurate presurgical planning. Changes in lesion diameter, volume, and enhancement are used to predict complete response, partial response, or nonresponse to therapy. However, residual disease may be overestimated or underestimated at MRI. Fibrosis, necrotic tumors, and residual benign masses may be causes of overestimation of residual disease. Nonmass lesions, invasive lobular carcinoma, hormone receptor-positive tumors, nonconcentric shrinkage patterns, the use of antiangiogenic therapy, and late-enhancing foci may be causes of underestimation of residual disease. In patients with known axillary lymph node metastasis, neoadjuvant therapy may be followed by targeted axillary dissection to avoid the potential morbidity associated with an axillary lymph node dissection. Diffusion-weighted imaging, radiomics, machine learning, and deep learning methods are under investigation to improve MRI accuracy in predicting treatment response.©RSNA, 2021.

Comparison of simultaneous multi-slice single-shot DWI to readout-segmented DWI for evaluation of breast lesions at 3T MRI.

PURPOSE: To compare diffusion-weighted imaging of the breast performed with a conventional readout-segmented echo-planar imaging (rs-EPI) sequence to when using a prototype simultaneous multi-slice single-shot EPI (SMS-ss-EPI) acquisition. METHOD: From September 2017 to December 2018, 26 women with histologically proven breast cancer were scanned with the conventional rs-EPI and the SMS-ss-EPI at 3 T during the same imaging examination. Four breast radiologists (4-13 years of experience) independently scored both acquired series of 25 women (one case was used for training) for overall image quality (1: extremely poor to 9: excellent) and artifacts (1: very disturbing to 5: not present). All lesions (n = 52; 40 malignant, 12 benign) were also evaluated for visibility (1: not visible, 2: visible if location is given, 3: visible). In addition, lesion characteristics were rated, and a BI-RADS score was given. Results were analyzed using visual grading characteristics and the resulting area under the curve (AUCVGC), weighted kappa, McNemar test, and dependent-samples t-test when appropriate. RESULTS: Overall, radiologists significantly preferred the image quality in rs-EPI over that of SMS-ss-EPI (AUCVGC: 0.698, P = 0.002). Infolding and ghosting, and distortion artifacts were significantly less apparent in the rs-EPI (AUCVGC: 0.660, P = 0.022 and AUCVGC: 0.700 P = 0.002, respectively). Lesions were, however, significantly better visible on the SMS-ss-EPI images (AUCVGC: 0.427, P = 0.016). Malignant lesions had significantly higher visibility with SMS-ss-EPI (P = 0.035). Sensitivity and specificity were comparable between both sequences (P = 0.760 and P = 0.549, respectively). CONCLUSIONS: Despite the perceived lower image quality and the increased presence of artifacts in the SMS-ss-EPI sequence, malignant lesions are better visualized using this sequence.

An interpretable classifier for high-resolution breast cancer screening images utilizing weakly supervised localization.

Medical images differ from natural images in significantly higher resolutions and smaller regions of interest. Because of these differences, neural network architectures that work well for natural images might not be applicable to medical image analysis. In this work, we propose a novel neural network model to address these unique properties of medical images. This model first uses a low-capacity, yet memory-efficient, network on the whole image to identify the most informative regions. It then applies another higher-capacity network to collect details from chosen regions. Finally, it employs a fusion module that aggregates global and local information to make a prediction. While existing methods often require lesion segmentation during training, our model is trained with only image-level labels and can generate pixel-level saliency maps indicating possible malignant findings. We apply the model to screening mammography interpretation: predicting the presence or absence of benign and malignant lesions. On the NYU Breast Cancer Screening Dataset, our model outperforms (AUC = 0.93) ResNet-34 and Faster R-CNN in classifying breasts with malignant findings. On the CBIS-DDSM dataset, our model achieves performance (AUC = 0.858) on par with state-of-the-art approaches. Compared to ResNet-34, our model is 4.1x faster for inference while using 78.4% less GPU memory. Furthermore, we demonstrate, in a reader study, that our model surpasses radiologist-level AUC by a margin of 0.11.

Magnetic Resonance Imaging in Screening of Breast Cancer.

Magnetic Resonance (MR) imaging is the most sensitive modality for breast cancer detection but is currently limited to screening women at high risk due to limited specificity and test accessibility. However, specificity of MR imaging improves with successive rounds of screening, and abbreviated approaches have the potential to increase access and decrease cost. There is growing evidence to support supplemental MR imaging in moderate-risk women, and current guidelines continue to evolve. Functional imaging has the potential to maximize survival benefit of screening. Leveraging MR imaging as a possible primary screening tool is therefore also being investigated in average-risk women.

Abbreviated MR Imaging for Breast Cancer.

Breast MR imaging is the most sensitive imaging method for the detection of breast cancer and detects more aggressive malignancies than mammography and ultrasound examination. Despite these advantages, breast MR imaging has low use rates for breast cancer screening. Abbreviated breast MR imaging, in which a limited number of breast imaging sequences are obtained, has been proposed as a way to solve cost and patient tolerance issues while preserving the high cancer detection rate of breast MR imaging. This review discusses abbreviated breast MR imaging, including protocols, multicenter clinical trial results, clinical workflow implementation challenges, and future directions.