Stand-Alone Use of Artificial Intelligence for Digital Mammography and Digital Breast Tomosynthesis Screening: A Retrospective Evaluation.

Background Use of artificial intelligence (AI) as a stand-alone reader for digital mammography (DM) or digital breast tomosynthesis (DBT) breast screening could ease radiologists' workload while maintaining quality. Purpose To retrospectively evaluate the stand-alone performance of an AI system as an independent reader of DM and DBT screening examinations. Materials and Methods Consecutive screening-paired and independently read DM and DBT images acquired between January 2015 and December 2016 were retrospectively collected from the Tomosynthesis Cordoba Screening Trial. An AI system computed a cancer risk score (range, 1-100) for DM and DBT examinations independently. AI stand-alone performance was measured using the area under the receiver operating characteristic curve (AUC) and sensitivity and recall rate at different operating points selected to have noninferior sensitivity compared with the human readings (noninferiority margin, 5%). The recall rate of AI and the human readings were compared using a McNemar test. Results A total of 15 999 DM and DBT examinations (113 breast cancers, including 98 screen-detected and 15 interval cancers) from 15 998 women (mean age, 58 years ± 6 [standard deviation]) were evaluated. AI achieved an AUC of 0.93 (95% CI: 0.89, 0.96) for DM and 0.94 (95% CI: 0.91, 0.97) for DBT. For DM, AI achieved noninferior sensitivity as a single (58.4%; 66 of 113; 95% CI: 49.2, 67.1) or double (67.3%; 76 of 113; 95% CI: 58.2, 75.2) reader, with a reduction in recall rate (P < .001) of up to 2% (95% CI: -2.4, -1.6). For DBT, AI achieved noninferior sensitivity as a single (77%; 87 of 113; 95% CI: 68.4, 83.8) or double (81.4%; 92 of 113; 95% CI: 73.3, 87.5) reader, but with a higher recall rate (P < .001) of up to 12.3% (95% CI: 11.7, 12.9). Conclusion Artificial intelligence could replace radiologists' readings in breast screening, achieving a noninferior sensitivity, with a lower recall rate for digital mammography but a higher recall rate for digital breast tomosynthesis. Published under a CC BY 4.0 license. See also the editorial by Fuchsjäger and Adelsmayr in this issue.

AI-based Strategies to Reduce Workload in Breast Cancer Screening with Mammography and Tomosynthesis: A Retrospective Evaluation.

Background The workflow of breast cancer screening programs could be improved given the high workload and the high number of false-positive and false-negative assessments. Purpose To evaluate if using an artificial intelligence (AI) system could reduce workload without reducing cancer detection in breast cancer screening with digital mammography (DM) or digital breast tomosynthesis (DBT). Materials and Methods Consecutive screening-paired and independently read DM and DBT images acquired from January 2015 to December 2016 were retrospectively collected from the Córdoba Tomosynthesis Screening Trial. The original reading settings were single or double reading of DM or DBT images. An AI system computed a cancer risk score for DM and DBT examinations independently. Each original setting was compared with a simulated autonomous AI triaging strategy (the least suspicious examinations for AI are not human-read; the rest are read in the same setting as the original, and examinations not recalled by radiologists but graded as very suspicious by AI are recalled) in terms of workload, sensitivity, and recall rate. The McNemar test with Bonferroni correction was used for statistical analysis. Results A total of 15 987 DM and DBT examinations (which included 98 screening-detected and 15 interval cancers) from 15 986 women (mean age ± standard deviation, 58 years ± 6) were evaluated. In comparison with double reading of DBT images (568 hours needed, 92 of 113 cancers detected, 706 recalls in 15 987 examinations), AI with DBT would result in 72.5% less workload (P < .001, 156 hours needed), noninferior sensitivity (95 of 113 cancers detected, P = .38), and 16.7% lower recall rate (P < .001, 588 recalls in 15 987 examinations). Similar results were obtained for AI with DM. In comparison with the original double reading of DM images (222 hours needed, 76 of 113 cancers detected, 807 recalls in 15 987 examinations), AI with DBT would result in 29.7% less workload (P < .001), 25.0% higher sensitivity (P < .001), and 27.1% lower recall rate (P < .001). Conclusion Digital mammography and digital breast tomosynthesis screening strategies based on artificial intelligence systems could reduce workload up to 70%. Published under a CC BY 4.0 license.

Impact of artificial intelligence support on accuracy and reading time in breast tomosynthesis image interpretation: a multi-reader multi-case study.

OBJECTIVES: Digital breast tomosynthesis (DBT) increases sensitivity of mammography and is increasingly implemented in breast cancer screening. However, the large volume of images increases the risk of reading errors and reading time. This study aims to investigate whether the accuracy of breast radiologists reading wide-angle DBT increases with the aid of an artificial intelligence (AI) support system. Also, the impact on reading time was assessed and the stand-alone performance of the AI system in the detection of malignancies was compared to the average radiologist. METHODS: A multi-reader multi-case study was performed with 240 bilateral DBT exams (71 breasts with cancer lesions, 70 breasts with benign findings, 339 normal breasts). Exams were interpreted by 18 radiologists, with and without AI support, providing cancer suspicion scores per breast. Using AI support, radiologists were shown examination-based and region-based cancer likelihood scores. Area under the receiver operating characteristic curve (AUC) and reading time per exam were compared between reading conditions using mixed-models analysis of variance. RESULTS: On average, the AUC was higher using AI support (0.863 vs 0.833; p = 0.0025). Using AI support, reading time per DBT exam was reduced (p < 0.001) from 41 (95% CI = 39-42 s) to 36 s (95% CI = 35- 37 s). The AUC of the stand-alone AI system was non-inferior to the AUC of the average radiologist (+0.007, p = 0.8115). CONCLUSIONS: Radiologists improved their cancer detection and reduced reading time when evaluating DBT examinations using an AI reading support system. KEY POINTS: • Radiologists improved their cancer detection accuracy in digital breast tomosynthesis (DBT) when using an AI system for support, while simultaneously reducing reading time. • The stand-alone breast cancer detection performance of an AI system is non-inferior to the average performance of radiologists for reading digital breast tomosynthesis exams. • The use of an AI support system could make advanced and more reliable imaging techniques more accessible and could allow for more cost-effective breast screening programs with DBT.

Amount of fibroglandular tissue FGT and background parenchymal enhancement BPE in relation to breast cancer risk and false positives in a breast MRI screening program : A retrospective cohort study.

OBJECTIVES: The purpose of this study is to evaluate the predictive value of the amount of fibroglandular tissue (FGT) and background parenchymal enhancement (BPE), measured at baseline on breast MRI, for breast cancer development and risk of false-positive findings in women at increased risk for breast cancer. METHODS: Negative baseline MRI scans of 1533 women participating in a screening program for women at increased risk for breast cancer between January 1, 2003, and January 1, 2014, were selected. Automated tools based on deep learning were used to obtain quantitative measures of FGT and BPE. Logistic regression using forward selection was used to assess relationships between FGT, BPE, cancer detection, false-positive recall, and false-positive biopsy. RESULTS: Sixty cancers were detected in follow-up. FGT was only associated to short-term cancer risk; BPE was not associated with cancer risk. High FGT and BPE did lead to more false-positive recalls at baseline (OR 1.259, p = 0.050, and OR 1.475, p = 0.003) and to more frequent false-positive biopsies at baseline (OR 1.315, p = 0.049, and OR 1.807, p = 0.002), but were not predictive for false-positive findings in subsequent screening rounds. CONCLUSIONS: FGT and BPE, measured on baseline MRI, are not predictive for overall breast cancer development in women at increased risk. High FGT and BPE lead to more false-positive findings at baseline. KEY POINTS: • Amount of fibroglandular tissue is only predictive for short-term breast cancer risk in women at increased risk. • Background parenchymal enhancement measured on baseline MRI is not predictive for breast cancer development in women at increased risk. • High amount of fibroglandular tissue and background parenchymal enhancement lead to more false-positive findings at baseline MRI.

Can we reduce the workload of mammographic screening by automatic identification of normal exams with artificial intelligence? A feasibility study.

PURPOSE: To study the feasibility of automatically identifying normal digital mammography (DM) exams with artificial intelligence (AI) to reduce the breast cancer screening reading workload. METHODS AND MATERIALS: A total of 2652 DM exams (653 cancer) and interpretations by 101 radiologists were gathered from nine previously performed multi-reader multi-case receiver operating characteristic (MRMC ROC) studies. An AI system was used to obtain a score between 1 and 10 for each exam, representing the likelihood of cancer present. Using all AI scores between 1 and 9 as possible thresholds, the exams were divided into groups of low- and high likelihood of cancer present. It was assumed that, under the pre-selection scenario, only the high-likelihood group would be read by radiologists, while all low-likelihood exams would be reported as normal. The area under the reader-averaged ROC curve (AUC) was calculated for the original evaluations and for the pre-selection scenarios and compared using a non-inferiority hypothesis. RESULTS: Setting the low/high-likelihood threshold at an AI score of 5 (high likelihood > 5) results in a trade-off of approximately halving (- 47%) the workload to be read by radiologists while excluding 7% of true-positive exams. Using an AI score of 2 as threshold yields a workload reduction of 17% while only excluding 1% of true-positive exams. Pre-selection did not change the average AUC of radiologists (inferior 95% CI > - 0.05) for any threshold except at the extreme AI score of 9. CONCLUSION: It is possible to automatically pre-select exams using AI to significantly reduce the breast cancer screening reading workload. KEY POINTS: • There is potential to use artificial intelligence to automatically reduce the breast cancer screening reading workload by excluding exams with a low likelihood of cancer. • The exclusion of exams with the lowest likelihood of cancer in screening might not change radiologists' breast cancer detection performance. • When excluding exams with the lowest likelihood of cancer, the decrease in true-positive recalls would be balanced by a simultaneous reduction in false-positive recalls.

The added value of mammography in different age-groups of women with and without BRCA mutation screened with breast MRI.

BACKGROUND: Breast magnetic resonance imaging (MRI) is the most sensitive imaging method for breast cancer detection and is therefore offered as a screening technique to women at increased risk of developing breast cancer. However, mammography is currently added from the age of 30 without proven benefits. The purpose of this study is to investigate the added cancer detection of mammography when breast MRI is available, focusing on the value in women with and without BRCA mutation, and in the age groups above and below 50 years. METHODS: This retrospective single-center study evaluated 6553 screening rounds in 2026 women at increased risk of breast cancer (1 January 2003 to 1 January 2014). Risk category (BRCA mutation versus others at increased risk of breast cancer), age at examination, recall, biopsy, and histopathological diagnosis were recorded. Cancer yield, false positive recall rate (FPR), and false positive biopsy rate (FPB) were calculated using generalized estimating equations for separate age categories (< 40, 40-50, 50-60, ≥ 60 years). Numbers of screens needed to detect an additional breast cancer with mammography (NSN) were calculated for the subgroups. RESULTS: Of a total of 125 screen-detected breast cancers, 112 were detected by MRI and 66 by mammography: 13 cancers were solely detected by mammography, including 8 cases of ductal carcinoma in situ. In BRCA mutation carriers, 3 of 61 cancers were detected only on mammography, while in other women 10 of 64 cases were detected with mammography alone. While 77% of mammography-detected-only cancers were detected in women ≥ 50 years of age, mammography also added more to the FPR in these women. Below 50 years the number of mammographic examinations needed to find an MRI-occult cancer was 1427. CONCLUSIONS: Mammography is of limited added value in terms of cancer detection when breast MRI is available for women of all ages who are at increased risk. While the benefit appears slightly larger in women over 50 years of age without BRCA mutation, there is also a substantial increase in false positive findings in these women.

Influence of Risk Category and Screening Round on the Performance of an MR Imaging and Mammography Screening Program in Carriers of the BRCA Mutation and Other Women at Increased Risk.

Purpose To evaluate the real-life performance of a breast cancer screening program for women with different categories of increased breast cancer risk with multiple follow-up rounds in an academic hospital with a large screening population. Materials and Methods Screening examinations (magnetic resonance [MR] imaging and mammography) for women at increased breast cancer risk (January 1, 2003, to January 1, 2014) were evaluated. Risk category, age, recall for workup of screening-detected abnormalities, biopsy, and histopathologic diagnosis were recorded. Recall rate, biopsy rate, positive predictive value of recall, positive predictive value of biopsy, cancer detection rate, sensitivity, and specificity were calculated for first and follow-up rounds. Results There were 8818 MR and 6245 mammographic examinations performed in 2463 women. Documented were 170 cancers; of these, there were 129 screening-detected cancers, 16 interval cancers, and 25 cancers discovered at prophylactic mastectomy. Overall sensitivity was 75.9% including the cancers discovered at prophylactic mastectomy (95% confidence interval: 69.5%, 82.4%) and 90.0% excluding those cancers (95% confidence interval: 83.3%, 93.7%). Sensitivity was lowest for carriers of the BRCA1 mutation (66.1% and 81.3% when including and not including cancers in prophylactic mastectomy specimens, respectively). Specificity was higher at follow-up (96.5%; 95% confidence interval: 96.0%, 96.9%) than in first rounds (85.1%; 95% confidence interval: 83.4%, 86.5%) and was high for both MR imaging (97.1%; 95% confidence interval: 96.7%, 97.5%) and mammography (98.7%; 95% confidence interval: 98.3%, 99.0%). Positive predictive value of recall and positive predictive value of biopsy were lowest in women who had only a family history of breast cancer. Conclusion Screening performance was dependent on risk category. Sensitivity was lowest in carriers of the BRCA1 mutation. The specificity of high-risk breast screening improved at follow-up rounds. © RSNA, 2017 Online supplemental material is available for this article.

One-view digital breast tomosynthesis as a stand-alone modality for breast cancer detection: do we need more?

PURPOSE: To compare the performance of one-view digital breast tomosynthesis (1v-DBT) to that of three other protocols combining DBT and mammography (DM) for breast cancer detection. MATERIALS AND METHODS: Six radiologists, three experienced with 1v-DBT in screening, retrospectively reviewed 181 cases (76 malignant, 50 benign, 55 normal) in two sessions. First, they scored sequentially: 1v-DBT (medio-lateral oblique, MLO), 1v-DBT (MLO) + 1v-DM (cranio-caudal, CC) and two-view DM + DBT (2v-DM+2v-DBT). The second session involved only 2v-DM. Lesions were scored using BI-RADS® and level of suspiciousness (1-10). Sensitivity, specificity, receiver operating characteristic (ROC) and jack-knife alternative free-response ROC (JAFROC) were computed. RESULTS: On average, 1v-DBT was non-inferior to any of the other protocols in terms of JAFROC figure-of-merit, area under ROC curve, sensitivity or specificity (p>0.391). While readers inexperienced with 1v-DBT screening improved their sensitivity when adding more images (69-79 %, p=0.019), experienced readers showed similar sensitivity (76 %) and specificity (70 %) between 1v-DBT and 2v-DM+2v-DBT (p=0.482). Subanalysis by lesion type and breast density showed no difference among modalities. CONCLUSION: Detection performance with 1v-DBT is not statistically inferior to 2v-DM or to 2v-DM+2v-DBT; its use as a stand-alone modality might be sufficient for readers experienced with this protocol. KEY POINTS: • One-view breast tomosynthesis is not inferior to two-view digital mammography. • One-view DBT is not inferior to 2-view DM plus 2-view DBT. • Training may lead to 1v-DBT being sufficient for screening.

New reconstruction algorithm for digital breast tomosynthesis: better image quality for humans and computers.

Background The image quality of digital breast tomosynthesis (DBT) volumes depends greatly on the reconstruction algorithm. Purpose To compare two DBT reconstruction algorithms used by the Siemens Mammomat Inspiration system, filtered back projection (FBP), and FBP with iterative optimizations (EMPIRE), using qualitative analysis by human readers and detection performance of machine learning algorithms. Material and Methods Visual grading analysis was performed by four readers specialized in breast imaging who scored 100 cases reconstructed with both algorithms (70 lesions). Scoring (5-point scale: 1 = poor to 5 = excellent quality) was performed on presence of noise and artifacts, visualization of skin-line and Cooper's ligaments, contrast, and image quality, and, when present, lesion visibility. In parallel, a three-dimensional deep-learning convolutional neural network (3D-CNN) was trained (n = 259 patients, 51 positives with BI-RADS 3, 4, or 5 calcifications) and tested (n = 46 patients, nine positives), separately with FBP and EMPIRE volumes, to discriminate between samples with and without calcifications. The partial area under the receiver operating characteristic curve (pAUC) of each 3D-CNN was used for comparison. Results EMPIRE reconstructions showed better contrast (3.23 vs. 3.10, P = 0.010), image quality (3.22 vs. 3.03, P < 0.001), visibility of calcifications (3.53 vs. 3.37, P = 0.053, significant for one reader), and fewer artifacts (3.26 vs. 2.97, P < 0.001). The 3D-CNN-EMPIRE had better performance than 3D-CNN-FBP (pAUC-EMPIRE = 0.880 vs. pAUC-FBP = 0.857; P < 0.001). Conclusion The new algorithm provides DBT volumes with better contrast and image quality, fewer artifacts, and improved visibility of calcifications for human observers, as well as improved detection performance with deep-learning algorithms.

Lesion Segmentation in Automated 3D Breast Ultrasound: Volumetric Analysis.

Mammography is the gold standard screening technique in breast cancer, but it has some limitations for women with dense breasts. In such cases, sonography is usually recommended as an additional imaging technique. A traditional sonogram produces a two-dimensional (2D) visualization of the breast and is highly operator dependent. Automated breast ultrasound (ABUS) has also been proposed to produce a full 3D scan of the breast automatically with reduced operator dependency, facilitating double reading and comparison with past exams. When using ABUS, lesion segmentation and tracking changes over time are challenging tasks, as the three-dimensional (3D) nature of the images makes the analysis difficult and tedious for radiologists. The goal of this work is to develop a semi-automatic framework for breast lesion segmentation in ABUS volumes which is based on the Watershed algorithm. The effect of different de-noising methods on segmentation is studied showing a significant impact ([Formula: see text]) on the performance using a dataset of 28 temporal pairs resulting in a total of 56 ABUS volumes. The volumetric analysis is also used to evaluate the performance of the developed framework. A mean Dice Similarity Coefficient of [Formula: see text] with a mean False Positive ratio [Formula: see text] has been obtained. The Pearson correlation coefficient between the segmented volumes and the corresponding ground truth volumes is [Formula: see text] ([Formula: see text]). Similar analysis, performed on 28 temporal (prior and current) pairs, resulted in a good correlation coefficient [Formula: see text] ([Formula: see text]) for prior and [Formula: see text] ([Formula: see text]) for current cases. The developed framework showed prospects to help radiologists to perform an assessment of ABUS lesion volumes, as well as to quantify volumetric changes during lesions diagnosis and follow-up.