AbstractObjectiveTo examine the accuracy of artificial intelligence (AI) for the detection of breast cancer in mammography screening practice.DesignSystematic review of test accuracy studies.Data ...sourcesMedline, Embase, Web of Science, and Cochrane Database of Systematic Reviews from 1 January 2010 to 17 May 2021.Eligibility criteriaStudies reporting test accuracy of AI algorithms, alone or in combination with radiologists, to detect cancer in women’s digital mammograms in screening practice, or in test sets. Reference standard was biopsy with histology or follow-up (for screen negative women). Outcomes included test accuracy and cancer type detected.Study selection and synthesisTwo reviewers independently assessed articles for inclusion and assessed the methodological quality of included studies using the QUality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool. A single reviewer extracted data, which were checked by a second reviewer. Narrative data synthesis was performed.ResultsTwelve studies totalling 131 822 screened women were included. No prospective studies measuring test accuracy of AI in screening practice were found. Studies were of poor methodological quality. Three retrospective studies compared AI systems with the clinical decisions of the original radiologist, including 79 910 women, of whom 1878 had screen detected cancer or interval cancer within 12 months of screening. Thirty four (94%) of 36 AI systems evaluated in these studies were less accurate than a single radiologist, and all were less accurate than consensus of two or more radiologists. Five smaller studies (1086 women, 520 cancers) at high risk of bias and low generalisability to the clinical context reported that all five evaluated AI systems (as standalone to replace radiologist or as a reader aid) were more accurate than a single radiologist reading a test set in the laboratory. In three studies, AI used for triage screened out 53%, 45%, and 50% of women at low risk but also 10%, 4%, and 0% of cancers detected by radiologists.ConclusionsCurrent evidence for AI does not yet allow judgement of its accuracy in breast cancer screening programmes, and it is unclear where on the clinical pathway AI might be of most benefit. AI systems are not sufficiently specific to replace radiologist double reading in screening programmes. Promising results in smaller studies are not replicated in larger studies. Prospective studies are required to measure the effect of AI in clinical practice. Such studies will require clear stopping rules to ensure that AI does not reduce programme specificity.Study registrationProtocol registered as PROSPERO CRD42020213590.
Screening mammography facilities closed during the COVID-19 pandemic in spring 2020. Recovery of screening volumes has varied across patient subgroups and facilities.
We compared screening ...mammography volumes and patient and facility characteristics between periods before COVID-19 and early and later postclosure recovery periods.
This retrospective study included screening mammograms performed in the same 2-month period (May 26-July 26) in 2019 (pre-COVID-19), 2020 (early recovery), and 2021 (late recovery after targeted interventions to expand access) and across multiple facility types (urban, suburban, community health center). Suburban sites had highest proportion of White patients and the greatest scheduling flexibility and expanded appointments during initial reopening. Findings were compared across years.
For White patients, volumes decreased 36.6% from 6550 in 2019 (4384 in 2020) and then increased 61.0% to 6579 in 2021; for patients with races other than White, volumes decreased 53.9% from 1321 in 2019 (609 in 2020) and then increased 136.8% to 1442 in 2021. The percentage of mammograms in patients with races other than White was 16.8% in 2019, 12.2% in 2020, and 18.0% in 2021. The proportion performed at the urban center was 55.3% in 2019, 42.2% in 2020, and 45.9% in 2021; the proportion at suburban sites was 34.0% in 2019, 49.2% in 2020, and 43.5% in 2021. Pre-COVID-19 volumes were reached by the sixth week after reopening for suburban sites but were not reached during early recovery for the other sites. The proportion that were performed on Saturday for suburban sites was similar across periods, whereas the proportion performed on Saturday for the urban site was 7.6% in 2019, 5.3% in 2020, and 8.8% in 2021; the community health center did not offer Saturday appointments during recovery.
After reopening, screening shifted from urban to suburban settings, with a disproportionate screening decrease in patients with races other than White. Initial delayed access at facilities serving underserved populations exacerbated disparities. Interventions to expand access resulted in late recovery volumes exceeding prepandemic volumes in patients with races other than White.
Interventions to support equitable access across facilities serving diverse patient populations may mitigate potential widening disparities in breast cancer diagnosis during the pandemic.
Breast cancer deaths averted over 3 decades Hendrick, R. Edward; Baker, Jay A.; Helvie, Mark A.
Cancer,
May 1, 2019, 2019-05-01, 2019-05-00, 20190501, Volume:
125, Issue:
9
Journal Article
Peer reviewed
Open access
Background
From 1975 to 1990, female breast cancer mortality rates in the United States increased by 0.4% per year. Since 1990, breast cancer mortality rates have fallen between 1.8% and 3.4% per ...year, a decrease that is attributed to increased mammography screening and improved treatment.
Methods
The authors used age‐adjusted female breast cancer mortality rate and population data from the Surveillance, Epidemiology, and End Results (SEER) program to estimate the number of breast cancer deaths averted by screening mammography and improved treatment since 1989. Four different assumptions regarding background mortality rates (in the absence of screening mammography and improved treatment) were used to estimate deaths averted for women aged 40 to 84 years by taking the difference between SEER‐reported mortality rates and background mortality rates for each 5‐year age group, multiplied by the population for each 5‐year age group. SEER data were used to estimate annual and cumulative breast cancer deaths averted in 2012 and 2015 and extrapolated SEER data were used to estimate deaths averted in 2018.
Results
The number of single‐year breast cancer deaths averted ranged from 20,860 to 33,842 in 2012, from 23,703 to 39,415 in 2015, and from 27,083 to 45,726 in 2018. Breast cancer mortality reductions ranged from 38.6% to 50.5% in 2012, from 41.5% to 54.2% in 2015, and from 45.3% to 58.3% in 2018. Cumulative breast cancer deaths averted since 1989 ranged from 237,234 to 370,402 in 2012, from 305,934 to 483,435 in 2015, and from 384,046 to 614,484 in 2018.
Conclusions
Since 1989, between 384,000 and 614,500 breast cancer deaths have been averted through the use of mammography screening and improved treatment.
Since the mid‐1980s, mammography screening has become widespread in the United States, and currently screens approximately one‐half of US women either annually or biennially. Since 1989, between 384,000 and 614,500 breast cancer deaths have been averted by mammography screening and improved breast cancer treatment.
Generative models are used as an alternative data augmentation technique to alleviate the data scarcity problem faced in the medical imaging field. Diffusion models have gathered special attention ...due to their innovative generation approach, the high quality of the generated images, and their relatively less complex training process compared with Generative Adversarial Networks. Still, the implementation of such models in the medical domain remains at an early stage. In this work, we propose exploring the use of diffusion models for the generation of high-quality, full-field digital mammograms using state-of-the-art conditional diffusion pipelines. Additionally, we propose using stable diffusion models for the inpainting of synthetic mass-like lesions on healthy mammograms. We introduce MAM-E, a pipeline of generative models for high-quality mammography synthesis controlled by a text prompt and capable of generating synthetic mass-like lesions on specific regions of the breast. Finally, we provide quantitative and qualitative assessment of the generated images and easy-to-use graphical user interfaces for mammography synthesis.
The purpose of this study was to compare contrast-enhanced spectral mammography (CESM) with mammography (MG) and combined CESM + MG in terms of detection and size estimation of histologically proven ...breast cancers in order to assess the potential to reduce radiation exposure. A total of 118 patients underwent MG and CESM and had final histological results. CESM was performed as a bilateral examination starting 2 min after injection of iodinated contrast medium. Three independent blinded radiologists read the CESM, MG, and CESM + MG images with an interval of at least 4 weeks to avoid case memorization. Sensitivity and size measurement correlation and differences were calculated, average glandular dose (AGD) levels were compared, and breast densities were reported. Fisher’s exact and Wilcoxon tests were performed. A total of 107 imaging pairs were available for analysis. Densities were ACR1: 2, ACR2: 45, ACR3: 42, and ACR4: 18. Mean AGD was 1.89 mGy for CESM alone, 1.78 mGy for MG, and 3.67 mGy for the combination. In very dense breasts, AGD of CESM was significantly lower than MG. Sensitivity across readers was 77.9 % for MG alone, 94.7 % for CESM, and 95 % for CESM + MG. Average tumor size measurement error compared to postsurgical pathology was −0.6 mm for MG, +0.6 mm for CESM, and +4.5 mm for CESM + MG (
p
< 0.001 for CESM + MG vs. both modalities). CESM alone has the same sensitivity and better size assessment as CESM + MG and was significantly better than MG with only 6.2 % increase in AGD. The combination of CESM + MG led to systematic size overestimation. When a CESM examination is planned, additional MG can be avoided, with the possibility of saving up to 61 % of radiation dose, especially in patients with dense breasts.
Abstract
Introduction/Objective
Mammographic documentation of microcalcifications requires biopsy of the microcalcifications with histologic study to demonstrate that the area of mammographic ...interest is examined histologically to exclude carcinoma. Initial levels may not demonstrate these microcalcifications requiring study of additional levels. Little data exists as to how deeply levels should be cut to document microcalcifications.
Methods/Case Report
A search of pathology records at the University of Missouri between January 1, 2022 and March 30, 2023 for reports containing the term “microcalcifications” in the clinical history or specimen designation was performed. For each identified case, all slides of blocks designated by the radiologist as containing microcalcifications were pulled for review. Our protocol was to cut initially two levels separated by 50 microns. If no microcalcifications were detected an additional 10 levels each separated by 100 microns were examined. The level at which microcalcifications were first detected was recorded.
Results (if a Case Study enter NA)
The search revealed four-hundred and thirty-one specimens meeting study criteria. Four-hundred and fifteen of these samples contained microcalcifications on histologic examination. Probability of finding microcalcifications in the initial level was 0.629 and 0.935 in the first six levels. The incremental cost per calcification was 1.5 for the first level and 66.7 for the sixth level. The relative cost per calcification in level 6 was 44.5. Levels beyond level 6 rarely disclose microcalcifications.
Conclusion
94% of microcalcifications documented by imaging were detected histologically in the first six levels using our protocol. The six-level protocol appears optimal for detection of microcalcifications. This study has significant implications for other specimen types. When a strong suspicion for a pathologic lesion is present on imaging but histologic sampling reveals no lesion, four further levels cut at 100 microns should be obtained.
While breast cancer continues to affect the lives of millions, contemporary writers and artists have responded to the ravages of the disease in creative expression. Mary K. DeShazer's book looks ...specifically at breast cancer memoirs and photographic narratives, a category she refers to as mammographies, signifying both the imaging technology by which most Western women discover they have this disease and the documentary imperatives that drive their written and visual accounts of it. Mammographies argues that breast cancer narratives of the past ten years differ from their predecessors in their bold address of previously neglected topics such as the link between cancer and environmental carcinogens, the ethics and efficacy of genetic testing and prophylactic mastectomy, and the shifting politics of prosthesis and reconstruction.
Objectives
Contrast-enhanced spectral mammography (CESM) is a promising problem-solving tool in women referred from a breast cancer screening program. We aimed to study the validity of preliminary ...results of CESM using a larger panel of radiologists with different levels of CESM experience.
Methods
All women referred from the Dutch breast cancer screening program were eligible for CESM. 199 consecutive cases were viewed by ten radiologists. Four had extensive CESM experience, three had no CESM experience but were experienced breast radiologists, and three were residents. All readers provided a BI-RADS score for the low-energy CESM images first, after which the score could be adjusted when viewing the entire CESM exam. BI-RADS 1-3 were considered benign and BI-RADS 4-5 malignant. With this cutoff, we calculated sensitivity, specificity and area under the ROC curve.
Results
CESM increased diagnostic accuracy in all readers. The performance for all readers using CESM was: sensitivity 96.9 % (+3.9 %), specificity 69.7 % (+33.8 %) and area under the ROC curve 0.833 (+0.188).
Conclusion
CESM is superior to conventional mammography, with excellent problem-solving capabilities in women referred from the breast cancer screening program. Previous results were confirmed even in a larger panel of readers with varying CESM experience.
Key Points
•
CESM is consistently superior to conventional mammography
•
CESM increases diagnostic accuracy regardless of a reader
’
s experience
•
CESM is an excellent problem
-
solving tool in recalls from screening programs