Artificial Intelligence Detection of Missed Cancers at Digital Mammography That Were Detected at Digital Breast Tomosynthesis.
Ontology highlight
ABSTRACT: Purpose
To investigate how an artificial intelligence (AI) system performs at digital mammography (DM) from a screening population with ground truth defined by digital breast tomosynthesis (DBT), and whether AI could detect breast cancers at DM that had originally only been detected at DBT.Materials and methods
In this secondary analysis of data from a prospective study, DM examinations from 14 768 women (mean age, 57 years), examined with both DM and DBT with independent double reading in the Malmӧ Breast Tomosynthesis Screening Trial (MBTST) (ClinicalTrials.gov: NCT01091545; data collection, 2010-2015), were analyzed with an AI system. Of 136 screening-detected cancers, 95 cancers were detected at DM and 41 cancers were detected only at DBT. The system identifies suspicious areas in the image, scored 1-100, and provides a risk score of 1 to 10 for the whole examination. A cancer was defined as AI detected if the cancer lesion was correctly localized and scored at least 62 (threshold determined by the AI system developers), therefore resulting in the highest examination risk score of 10. Data were analyzed with descriptive statistics, and detection performance was analyzed with receiver operating characteristics.Results
The highest examination risk score was assigned to 10% (1493 of 14 786) of the examinations. With 90.8% specificity, the AI system detected 75% (71 of 95) of the DM-detected cancers and 44% (18 of 41) of cancers at DM that had originally been detected only at DBT. The majority were invasive cancers (17 of 18).Conclusion
Almost half of the additional DBT-only screening-detected cancers in the MBTST were detected at DM with AI. AI did not reach double reading performance; however, if combined with double reading, AI has the potential to achieve a substantial portion of the benefit of DBT screening.Keywords: Computer-aided Diagnosis, Mammography, Breast, Diagnosis, Classification, Application DomainClinical trial registration no. NCT01091545© RSNA, 2021.
SUBMITTER: Dahlblom V
PROVIDER: S-EPMC8637235 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Although computer-aided diagnosis (CAD) is widely used in mammography, conventional CAD programs that use prompts to indicate potential cancers on the mammograms have not led to an improvement in diagnostic accuracy. Because of the advances in machine learning, especially with use of deep (multilayered) convolutional neural networks, artificial intelligence has undergone a transformation that has improved the quality of the predictions of the models. Recently, such deep learning algorithms have been applied to mammography and digital breast tomosynthesis (DBT). In this review, the authors explain how deep learning works in the context of mammography and DBT and define the important technical challenges. Subsequently, they discuss the current status and future perspectives of artificial intelligence-based clinical applications for mammography, DBT, and radiomics. Available algorithms are advanced and approach the performance of radiologists-especially for cancer detection and risk prediction at mammography. However, clinical validation is largely lacking, and it is not clear how the power of deep learning should be used to optimize practice. Further development of deep learning models is necessary for DBT, and this requires collection of larger databases. It is expected that deep learning will eventually have an important role in DBT, including the generation of synthetic images.
| S-EPMC6822772 | biostudies-literature
Project description:To determine whether cmAssist™, an artificial intelligence-based computer-aided detection (AI-CAD) algorithm, can be used to improve radiologists' sensitivity in breast cancer screening and detection. A blinded retrospective study was performed with a panel of seven radiologists using a cancer-enriched data set from 122 patients that included 90 false-negative mammograms obtained up to 5.8 years prior to diagnosis and 32 BIRADS 1 and 2 patients with a 2-year follow-up of negative diagnosis. The mammograms were performed between February 7, 2008 (earliest) and January 8, 2016 (latest), and were all originally interpreted as negative in conjunction with R2 ImageChecker CAD, version 10.0. In this study, the readers analyzed the 122 studies before and after review of cmAssist™, an AI-CAD software for mammography. The statistical significance of our findings was evaluated using Student's t test and bootstrap statistical analysis. There was a substantial and significant improvement in radiologist accuracy with use of cmAssist, as demonstrated in the 7.2% increase in the area-under-the-curve (AUC) of the receiver operating characteristic (ROC) curve with two-sided p value?<?0.01 for the reader group. All radiologists showed a significant improvement in their cancer detection rate (CDR) with the use of cmAssist (two-sided p value?=?0.030, confidence interval?=?95%). The readers detected between 25 and 71% (mean 51%) of the early cancers without assistance. With cmAssist, the overall reader CDR was 41 to 76% (mean 62%). The percentage increase in CDR for the reader panel was significant, ranging from 6 to 64% (mean 27%) with the use of cmAssist. There was less than 1% increase in the readers' false-positive recalls with use of cmAssist. With the use of cmAssist TM, there was a substantial and statistically significant improvement in radiologists' accuracy and sensitivity for detection of cancers that were originally missed. The percentage increase in CDR for the radiologists in the reader panel ranged from 6 to 64% (mean 27%) with the use of cmAssist, with negligible increase in false-positive recalls.
| S-EPMC6646649 | biostudies-literature
Project description:ObjectivesDigital breast tomosynthesis (DBT) can detect more cancers than the current standard breast screening method, digital mammography (DM); however, it can substantially increase the reading workload and thus hinder implementation in screening. Artificial intelligence (AI) might be a solution. The aim of this study was to retrospectively test different ways of using AI in a screening workflow.MethodsAn AI system was used to analyse 14,772 double-read single-view DBT examinations from a screening trial with paired DM double reading. Three scenarios were studied: if AI can identify normal cases that can be excluded from human reading; if AI can replace the second reader; if AI can replace both readers. The number of detected cancers and false positives was compared with DM or DBT double reading.ResultsBy excluding normal cases and only reading 50.5% (7460/14,772) of all examinations, 95% (121/127) of the DBT double reading detected cancers could be detected. Compared to DM screening, 27% (26/95) more cancers could be detected (p < 0.001) while keeping recall rates at the same level. With AI replacing the second reader, 95% (120/127) of the DBT double reading detected cancers could be detected-26% (25/95) more than DM screening (p < 0.001)-while increasing recall rates by 53%. AI alone with DBT has a sensitivity similar to DM double reading (p = 0.689).ConclusionAI can open up possibilities for implementing DBT screening and detecting more cancers with the total reading workload unchanged. Considering the potential legal and psychological implications, replacing the second reader with AI would probably be most the feasible approach.Key points• Breast cancer screening with digital breast tomosynthesis and artificial intelligence can detect more cancers than mammography screening without increasing screen-reading workload. • Artificial intelligence can either exclude low-risk cases from double reading or replace the second reader. • Retrospective study based on paired mammography and digital breast tomosynthesis screening data.
| S-EPMC10121528 | biostudies-literature
Project description:ImportanceBreast cancer screening examinations using digital breast tomosynthesis (DBT) has been shown to be associated with decreased false-positive test results and increased breast cancer detection compared with digital mammography (DM). Little is known regarding the size and stage of breast cancer types detected and their association with age and breast density.ObjectiveTo determine whether screening examinations using DBT detect breast cancers that are associated with an improved prognosis and to compare the detection rates by patient age and breast density.Design, setting, and participantsThis retrospective analysis of prospective cohort data from 3 research centers in the Population-based Research Optimizing Screening Through Personalized Regimens (PROSPR) consortium included data of women aged 40 to 74 years who underwent screening examinations using DM and DBT from January 1, 2011, through September 30, 2014. Statistical analysis was performed from November 8, 2017, to August 14, 2018.ExposuresUse of DBT as a supplement to DM at breast cancer screening examination.Main outcomes and measuresRecall rate, cancer detection rate, positive predictive value, biopsy rate, and distribution of invasive cancer subtypes.ResultsAmong 96 269 women (mean [SD] patient age for all examinations, 55.9 [9.0] years), patient age was 56.4 (9.0) years for DM and 54.6 (8.9) years for DBT. Of 180 340 breast cancer screening examinations, 129 369 examinations (71.7%) used DM and 50 971 examinations (28.3%) used DBT. Screening examination with DBT (73 of 99 women [73.7%]) was associated with the detection of smaller, more often node-negative, HER2-negative, invasive cancers compared with DM (276 of 422 women [65.4%]). Screening examination with DBT was also associated with lower recall (odds ratio, 0.64; 95% CI, 0.57-0.72; P < .001) and higher cancer detection (odds ratio, 1.41; 95% CI, 1.05-1.89; P = .02) compared with DM for all age groups even when stratified by breast density. The largest increase in cancer detection rate and the greatest shift toward smaller, node-negative invasive cancers detected with DBT was for women aged 40 to 49 years. For women aged 40 to 49 years with nondense breasts, the cancer detection rate for examinations using DBT was 1.70 per 1000 women higher compared with the rate using DM; for women with dense breasts, the cancer detection rate was 2.27 per 1000 women higher for DBT. For these younger women, screening with DBT was associated with only 7 of 28 breast cancers (25.0%) categorized as poor prognosis compared with 19 of 47 breast cancers (40.4%) when screening with DM.Conclusions and relevanceThe findings suggest that screening with DBT is associated with increased specificity and an increased proportion of breast cancers detected with better prognosis compared with DM. In the subgroup of women aged 40 to 49 years, routine DBT screening may have a favorable risk-benefit ratio.
| S-EPMC6512257 | biostudies-literature
Project description:BackgroundWe examined whether digital breast tomosynthesis (DBT) detects differentially in high- or low-density screens.MethodsWe searched six databases (2009-2020) for studies comparing DBT and digital mammography (DM), and reporting cancer detection rate (CDR) and/or recall rate by breast density. Meta-analysis was performed to pool incremental CDR and recall rate for DBT (versus DM) for high- and low-density (dichotomised based on BI-RADS) and within-study differences in incremental estimates between high- and low-density. Screening settings (European/US) were compared.ResultsPooled within-study difference in incremental CDR for high- versus low-density was 1.0/1000 screens (95% CI: 0.3, 1.6; p = 0.003). Estimates were not significantly different in US (0.6/1000; 95% CI: 0.0, 1.3; p = 0.05) and European (1.9/1000; 95% CI: 0.3, 3.5; p = 0.02) settings (p for subgroup difference = 0.15). For incremental recall rate, within-study differences between density subgroups differed by setting (p < 0.001). Pooled incremental recall was less in high- versus low-density screens (-0.9%; 95% CI: -1.4%, -0.4%; p < 0.001) in US screening, and greater (0.8%; 95% CI: 0.3%, 1.3%; p = 0.001) in European screening.ConclusionsDBT has differential incremental cancer detection and recall by breast density. Although incremental CDR is greater in high-density, a substantial proportion of additional cancers is likely to be detected in low-density screens. Our findings may assist screening programmes considering DBT for density-tailored screening.
| S-EPMC9276736 | biostudies-literature
Project description:Background Breast Imaging Reporting and Data System (BI-RADS) breast density categories assigned by interpreting radiologists often influence decisions surrounding supplemental breast cancer screening and risk assessment. The landscape of mammographic screening continuously evolves, and different mammographic screening modalities may result in different perception of density, reflected in different assignment of BI-RADS density categories. Purpose To investigate the effect of screening mammography modality on BI-RADS breast density assessments. Materials and Methods Data were retrospectively analyzed from 24 736 individual women (42.3% [10 455 of 24 736] white women, 57.7% [14 281 of 24 736] black women; mean age, 56.3 years; age range, 40.0-74.9 years) who underwent from one to seven mammographic screening examinations from September 2010 through February 2017 (60 766 examinations). Three screening modalities were used: digital mammography alone (8935 examinations); digital mammography with digital breast tomosynthesis (DBT; 30 779 examinations); and synthetic mammography with DBT (21 052 examinations). Random-effects logistic regression analysis was performed to estimate the likelihood of assignment to high versus low BI-RADS density category according to each modality, adjusted for ethnicity, age, body mass index (BMI), and radiologist. The interactions of modality with ethnicity and BMI on density categorization were also tested with the model. Results Women screened with DBT versus digital mammography alone had lower likelihood regarding categorization of high density breasts (digital mammography and DBT vs digital mammography: odds ratio, 0.69 [95% confidence interval: 0.61, 0.80], P < .001; synthetic mammography and DBT vs digital mammography: odds ratio, 0.43 [95% confidence interval: 0.37, 0.50], P < .001). Lower likelihood of high density was also observed at synthetic mammography and DBT compared with digital mammography and DBT (odds ratio, 0.62; 95% confidence interval: 0.56, 0.69; P < .001). There were interactions of modality with ethnicity (P = .007) and BMI (P = .003) on breast density assessment, with greater differences in density categorization according to modality observed for black women than for white women and groups with higher BMI. Conclusion Breast density categorization may vary by screening mammographic modality, and this effect appears to vary by ethnicity and body mass index. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Philpotts in this issue.
| S-EPMC6493215 | biostudies-literature
Project description:ImportanceMany US radiologists have screening mammography recall rates above the expert-recommended threshold of 12%. The influence of digital breast tomosynthesis (DBT) on the distribution of radiologist recall rates is uncertain.ObjectiveTo evaluate radiologists' recall and cancer detection rates before and after beginning interpretation of DBT examinations.Design, setting, and participantsThis cohort study included 198 radiologists from 104 radiology facilities in the Breast Cancer Surveillance Consortium who interpreted 251?384 DBT and 2?000?681 digital mammography (DM) screening examinations from 2009 to 2017, including 126 radiologists (63.6%) who interpreted DBT examinations during the study period and 72 (36.4%) who exclusively interpreted DM examinations (to adjust for secular trends). Data were analyzed from April 2018 to July 2019.ExposuresDigital breast tomosynthesis and DM screening examinations.Main outcomes and measuresRecall rate and cancer detection rate.ResultsA total of 198 radiologists interpreted 2?252?065 DM and DBT examinations (2?000?681 [88.8%] DM examinations; 251?384 [11.2%] DBT examinations; 710?934 patients [31.6%] aged 50-59 years; 1?448?981 [64.3%] non-Hispanic white). Among the 126 radiologists (63.6%) who interpreted DBT examinations, 83 (65.9%) had unadjusted DM recall rates of no more than 12% before using DBT, with a median (interquartile range) recall rate of 10.0% (7.5%-13.0%). On DBT examinations, 96 (76.2%) had an unadjusted recall rate of no more than 12%, with a median (interquartile range) recall rate of 8.8% (6.3%-11.3%). A secular trend in recall rate was observed, with the multivariable-adjusted risk of recall on screening examinations declining by 1.2% (95% CI, 0.9%-1.5%) per year. After adjusting for examination characteristics and secular trends, recall rates were 15% lower on DBT examinations compared with DM examinations interpreted before DBT use (relative risk, 0.85; 95% CI, 0.83-0.87). Adjusted recall rates were significantly lower on DBT examinations compared with DM examinations interpreted before DBT use for 45 radiologists (35.7%) and significantly higher for 18 (14.3%); 63 (50.0%) had no statistically significant change. The unadjusted cancer detection rate on DBT was 5.3 per 1000 examinations (95% CI, 5.0-5.7 per 1000 examinations) compared with 4.7 per 1000 examinations (95% CI, 4.6-4.8 per 1000 examinations) on DM examinations interpreted before DM use (multivariable-adjusted risk ratio, 1.21; 95% CI, 1.11-1.33).Conclusions and relevanceIn this study, DBT was associated with an overall decrease in recall rate and an increase in cancer detection rate. However, our results indicated that there is wide variability among radiologists, including a subset of radiologists who experienced increased recall rates on DBT examinations. Radiology practices should audit radiologist DBT screening performance and consider additional DBT training for radiologists whose performance does not improve as expected.
| S-EPMC7292996 | biostudies-literature
Project description:BackgroundOur objective was to perform a systematic review and meta-analysis comparing the breast cancer detection rate (CDR), invasive CDR, recall rate, and positive predictive value 1 (PPV1) of digital mammography (DM) alone, combined digital breast tomosynthesis (DBT) and DM, combined DBT and synthetic 2-dimensional mammography (S2D), and DBT alone.MethodsMEDLINE and Embase were searched until April 2020 to identify comparative design studies reporting on patients undergoing routine breast cancer screening. Random effects model proportional meta-analyses estimated CDR, invasive CDR, recall rate, and PPV1. Meta-regression modeling was used to compare imaging modalities. All statistical tests were 2-sided.ResultsForty-two studies reporting on 2 606 296 patients (13 003 breast cancer cases) were included. CDR was highest in combined DBT and DM (6.36 per 1000 screened, 95% confidence interval [CI] = 5.62 to 7.14, P < .001), and combined DBT and S2D (7.40 per 1000 screened, 95% CI = 6.49 to 8.37, P < .001) compared with DM alone (4.68 per 1000 screened, 95% CI = 4.28 to 5.11). Invasive CDR was highest in combined DBT and DM (4.53 per 1000 screened, 95% CI = 3.97 to 5.12, P = .003) and combined DBT and S2D (5.68 per 1000 screened, 95% CI = 4.43 to 7.09, P < .001) compared with DM alone (3.42 per 1000 screened, 95% CI = 3.02 to 3.83). Recall rate was lowest in combined DBT and S2D (42.3 per 1000 screened, 95% CI = 37.4 to 60.4, P<.001). PPV1 was highest in combined DBT and DM (10.0%, 95% CI = 8.0% to 12.0%, P = .004), and combined DBT and S2D (16.0%, 95% CI = 10.0% to 23.0%, P < .001), whereas no difference was detected for DBT alone (7.0%, 95% CI = 6.0% to 8.0%, P = .75) compared with DM alone (7.0%, 95.0% CI = 5.0% to 8.0%).ConclusionsOur findings provide evidence on key performance metrics for DM, DBT alone, combined DBT and DM, and combined DBT and S2D, which may inform optimal application of these modalities for breast cancer screening.
| S-EPMC8168096 | biostudies-literature
Project description:BackgroundDigital breast tomosynthesis (DBT) has become a prevalent mode of breast cancer screening in recent years. Although older women are commonly screened for breast cancer, little is known about screening outcomes using DBT among older women.ObjectiveTo assess proximal screening outcomes with DBT compared to traditional two-dimensional(2-D) mammography among women 67-74 and women 75 and older.DesignCohort study.ParticipantsMedicare fee-for-service beneficiaries aged 67 years and older with no history of prior cancer who received a screening mammogram in 2015.Main measuresUse of subsequent imaging (ultrasound and diagnostic mammography) as an indication of recall, breast cancer detection, and characteristics of breast cancer at the time of diagnosis. Analyses used weighted logistic regression to adjust for potential confounders.Key resultsOur study included 26,406 women aged 67-74 and 17,001 women 75 and older who were screened for breast cancer. Among women 75 and older, the rate of subsequent imaging among women screened with DBT did not differ significantly from 2-D mammography (91.8 versus 97.0 per 1,000 screening mammograms, p=0.37). In this age group, DBT was associated with 2.1 additional cancers detected per 1,000 screening mammograms compared to 2D (11.5 versus 9.4, p=0.003), though these additional cancers were almost exclusively in situ and stage I invasive cancers. For women 67-74 years old, DBT was associated with a higher rate of subsequent imaging than 2-D mammography (113.9 versus 100.3, p=0.004) and a higher rate of stage I invasive cancer detection (4.7 versus 3.7, p=0.002), but not other stages.ConclusionsBreast cancer screening with DBT was not associated with lower rates of subsequent imaging among older women. Most additional cancers detected with DBT were early stage. Whether detecting these additional early-stage cancers among older women improves health outcomes remains uncertain.
| S-EPMC8483166 | biostudies-literature
Project description:ImportanceAn accurate and robust artificial intelligence (AI) algorithm for detecting cancer in digital breast tomosynthesis (DBT) could significantly improve detection accuracy and reduce health care costs worldwide.ObjectivesTo make training and evaluation data for the development of AI algorithms for DBT analysis available, to develop well-defined benchmarks, and to create publicly available code for existing methods.Design, setting, and participantsThis diagnostic study is based on a multi-institutional international grand challenge in which research teams developed algorithms to detect lesions in DBT. A data set of 22 032 reconstructed DBT volumes was made available to research teams. Phase 1, in which teams were provided 700 scans from the training set, 120 from the validation set, and 180 from the test set, took place from December 2020 to January 2021, and phase 2, in which teams were given the full data set, took place from May to July 2021.Main outcomes and measuresThe overall performance was evaluated by mean sensitivity for biopsied lesions using only DBT volumes with biopsied lesions; ties were broken by including all DBT volumes.ResultsA total of 8 teams participated in the challenge. The team with the highest mean sensitivity for biopsied lesions was the NYU B-Team, with 0.957 (95% CI, 0.924-0.984), and the second-place team, ZeDuS, had a mean sensitivity of 0.926 (95% CI, 0.881-0.964). When the results were aggregated, the mean sensitivity for all submitted algorithms was 0.879; for only those who participated in phase 2, it was 0.926.Conclusions and relevanceIn this diagnostic study, an international competition produced algorithms with high sensitivity for using AI to detect lesions on DBT images. A standardized performance benchmark for the detection task using publicly available clinical imaging data was released, with detailed descriptions and analyses of submitted algorithms accompanied by a public release of their predictions and code for selected methods. These resources will serve as a foundation for future research on computer-assisted diagnosis methods for DBT, significantly lowering the barrier of entry for new researchers.
| S-EPMC9951043 | biostudies-literature