Artificial Intelligence (AI) for Screening Mammography, From the AJR Special Series on AI Applications

Mammography-based artificial intelligence for breast cancer detection, diagnosis, and BI-RADS categorization using multi-view and multi-level convolutional neural networks

PURPOSE

We developed an artificial intelligence system (AIS) using multi-view multi-level convolutional neural networks for breast cancer detection, diagnosis, and BI-RADS categorization support in mammography.

METHODS

Twenty-four thousand eight hundred sixty-six breasts from 12,433 Asian women between August 2012 and December 2018 were enrolled. The study consisted of three parts: (1) evaluation of AIS performance in malignancy diagnosis; (2) stratified analysis of BI-RADS 3-4 subgroups with AIS; and (3) reassessment of BI-RADS 0 breasts with AIS assistance. We further evaluate AIS by conducting a counterbalance-designed AI-assisted study, where ten radiologists read 1302 cases with/without AIS assistance. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, and F1 score were measured.

RESULTS

The AIS yielded AUC values of 0.995, 0.933, and 0.947 for malignancy diagnosis in the validation set, testing set 1, and testing set 2, respectively. Within BI-RADS 3-4 subgroups with pathological results, AIS downgraded 83.1% of false-positives into benign groups, and upgraded 54.1% of false-negatives into malignant groups. AIS also successfully assisted radiologists in identifying 7 out of 43 malignancies initially diagnosed with BI-RADS 0, with a specificity of 96.7%. In the counterbalance-designed AI-assisted study, the average AUC across ten readers significantly improved with AIS assistance (p = 0.001).

CONCLUSION

AIS can accurately detect and diagnose breast cancer on mammography and further serve as a supportive tool for BI-RADS categorization.

CRITICAL RELEVANCE STATEMENT

An AI risk assessment tool employing deep learning algorithms was developed and validated for enhancing breast cancer diagnosis from mammograms, to improve risk stratification accuracy, particularly in patients with dense breasts, and serve as a decision support aid for radiologists.

KEY POINTS

The false positive and negative rates of mammography diagnosis remain high. The AIS can yield a high AUC for malignancy diagnosis. The AIS is important in stratifying BI-RADS categorization.

Challenges in Implementing Artificial Intelligence in Breast Cancer Screening Programs: Systematic Review and Framework for Safe Adoption

BACKGROUND

Artificial intelligence (AI) studies show promise in enhancing accuracy and efficiency in mammographic screening programs worldwide. However, its integration into clinical workflows faces several challenges, including unintended errors, the need for professional training, and ethical concerns. Notably, specific frameworks for AI imaging in breast cancer screening are still lacking.

OBJECTIVE

This study aims to identify the challenges associated with implementing AI in breast screening programs and to apply the Consolidated Framework for Implementation Research (CFIR) to discuss a practical governance framework for AI in this context.

METHODS

Three electronic databases (PubMed, Embase, and MEDLINE) were searched using combinations of the keywords "artificial intelligence," "regulation," "governance," "breast cancer," and "screening." Original studies evaluating AI in breast cancer detection or discussing challenges related to AI implementation in this setting were eligible for review. Findings were narratively synthesized and subsequently mapped directly onto the constructs within the CFIR.

RESULTS

A total of 1240 results were retrieved, with 20 original studies ultimately included in this systematic review. The majority (n=19) focused on AI-enhanced mammography, while 1 addressed AI-enhanced ultrasound for women with dense breasts. Most studies originated from the United States (n=5) and the United Kingdom (n=4), with publication years ranging from 2019 to 2023. The quality of papers was rated as moderate to high. The key challenges identified were reproducibility, evidentiary standards, technological concerns, trust issues, as well as ethical, legal, societal concerns, and postadoption uncertainty. By aligning these findings with the CFIR constructs, action plans targeting the main challenges were incorporated into the framework, facilitating a structured approach to addressing these issues.

CONCLUSIONS

This systematic review identifies key challenges in implementing AI in breast cancer screening, emphasizing the need for consistency, robust evidentiary standards, technological advancements, user trust, ethical frameworks, legal safeguards, and societal benefits. These findings can serve as a blueprint for policy makers, clinicians, and AI developers to collaboratively advance AI adoption in breast cancer screening.

TRIAL REGISTRATION

PROSPERO CRD42024553889; https://tinyurl.com/mu4nwcxt.

The impact of AI feedback on the accuracy of diagnosis, decision switching and trust in radiography

Artificial intelligence decision support systems have been proposed to assist a struggling National Health Service (NHS) workforce in the United Kingdom. Its implementation in UK healthcare systems has been identified as a priority for deployment. Few studies have investigated the impact of the feedback from such systems on the end user. This study investigated the impact of two forms of AI feedback (saliency/heatmaps and AI diagnosis with percentage confidence) on student and qualified diagnostic radiographers' accuracy when determining binary diagnosis on skeletal radiographs. The AI feedback proved beneficial to accuracy in all cases except when the AI was incorrect and for pathological cases in the student group. The self-reported trust of all participants decreased from the beginning to the end of the study. The findings of this study should guide developers in the provision of the most advantageous forms of AI feedback and direct educators in tailoring education to highlight weaknesses in human interaction with AI-based clinical decision support systems.

The Transformative Power of Digital Breast Tomosynthesis and Artificial Intelligence in Breast Cancer Diagnosis

The integration of Digital Breast Tomosynthesis (DBT) and Artificial Intelligence (AI) represents a significant advance in breast cancer screening. This combination aims to address several challenges inherent in traditional screening while promising an improvement in healthcare delivery across multiple dimensions. For patients, this technological synergy has the potential to lower the number of unnecessary recalls and associated procedures such as biopsies, thereby reducing patient anxiety and improving overall experience without compromising diagnostic accuracy. For radiologists, the use of combined AI and DBT could significantly decrease workload and reduce fatigue by effectively highlighting breast imaging abnormalities, which is especially beneficial in high-volume clinical settings. Health systems stand to gain from streamlined workflows and the facilitated deployment of DBT, which is particularly valuable in areas with a scarcity of specialized breast radiologists. However, despite these potential benefits, substantial challenges remain. Bridging the gap between the development of complex AI algorithms and implementation into clinical practice requires ongoing research and development. This is essential to optimize the reliability of these systems and ensure they are accessible to healthcare providers and patients, who are the ultimate beneficiaries of this technological advancement. This article reviews the benefits of combined AI-DBT imaging, particularly the ability of AI to enhance the benefits of DBT and reduce its existing limitations.

Breast Cancer Detection with Standalone AI versus Radiologist Interpretation of Unilateral Surveillance Mammography after Mastectomy

Background Limited data are available regarding the accuracy of artificial intelligence (AI) algorithms trained on bilateral mammograms for second breast cancer surveillance in patients with a personal history of breast cancer treated with unilateral mastectomy. Purpose To compare the performance of standalone AI for second breast cancer surveillance on unilateral mammograms with that of radiologists reading mammograms without AI assistance. Materials and Methods In this retrospective institutional database study, patients who were diagnosed with breast cancer between January 2001 and December 2018 and underwent postmastectomy surveillance mammography from January 2011 to March 2023 were included. Radiologists' mammogram interpretations without AI assistance were collected from these records and compared with AI interpretations of the same mammograms. The reference standards were histologic examination and 1-year follow-up data. The cancer detection rate per 1000 screening examinations, sensitivity, and specificity of standalone AI and the radiologists' interpretations without AI were compared using the McNemar test. Results Among the 4184 asymptomatic female patients (mean age, 52 years), 111 (2.7%) had contralateral second breast cancer. The cancer detection rate (17.4 per 1000 examinations [73 of 4184]; 95% CI: 13.7, 21.9) and sensitivity (65.8% [73 of 111]; 95% CI: 56.2, 74.5) were greater for standalone AI than for radiologists (14.6 per 1000 examinations [61 of 4184]; 95% CI: 11.2, 18.7; P = .01; 55.0% [61 of 111]; 95% CI: 45.2, 64.4; P = .01). The specificity was lower for standalone AI than for radiologists (91.5% [3725 of 4073]; 95% CI: 90.6, 92.3 vs 98.1% [3996 of 4073]; 95% CI: 97.6, 98.5; P < .001). AI detected 16 of 50 (32%) cancers missed by radiologists; however, 34 of 111 (30.6%) breast cancers were missed by both radiologists and AI. Conclusion Standalone AI for surveillance mammography showed higher sensitivity with lower specificity for contralateral breast cancer detection in patients treated with unilateral mastectomy than radiologists without AI assistance. © RSNA, 2025 Supplemental material is available for this article. See also the editorial by Philpotts in this issue.

Traditional versus modern approaches to screening mammography: a comparison of computer-assisted detection for synthetic 2D mammography versus an artificial intelligence algorithm for digital breast tomosynthesis

PURPOSE

Traditional computer-assisted detection (CADe) algorithms were developed for 2D mammography, while modern artificial intelligence (AI) algorithms can be applied to 2D mammography and/or digital breast tomosynthesis (DBT). The objective is to compare the performance of a traditional machine learning CADe algorithm for synthetic 2D mammography to a deep learning-based AI algorithm for DBT on the same mammograms.

METHODS

Mammographic examinations from 764 patients (mean age 58 years ± 11) with 106 biopsy-proven cancers and 658 cancer-negative cases were analyzed by a CADe algorithm (ImageChecker v10.0, Hologic, Inc.) and an AI algorithm (Genius AI Detection v2.0, Hologic, Inc.). Synthetic 2D images were used for CADe analysis, and DBT images were used for AI analysis. For each algorithm, an overall case score was defined as the highest score of all lesion marks, which was used to determine the area under the receiver operating characteristic curve (AUC).

RESULTS

The overall AUC was higher for 3D AI than 2D CADe (0.873 versus 0.693, P < 0.001). Lesion-specific sensitivity of 3D AI was higher than 2D CADe (94.3 versus 72.6%, P = 0.002). Specificity of 3D AI was higher than 2D CADe (54.3 versus 16.7%, P < 0.001), and the rate of false marks on non-cancer cases was lower for 3D AI than 2D CADe (0.91 versus 3.24 per exam, P < 0.001).

CONCLUSION

A deep learning-based AI algorithm applied to DBT images significantly outperformed a traditional machine learning CADe algorithm applied to synthetic 2D mammographic images, with regard to AUC, sensitivity, and specificity.

Evaluation of a Multi-Instant Multimodal Artificial Intelligence System Supporting Interpretive and Noninterpretive Functions

OBJECTIVE

Artificial intelligence (AI) has been shown to hold promise for improving breast cancer screening, offering advanced capabilities to enhance diagnostic accuracy and efficiency. This study aimed to evaluate the impact of a multimodal multi-instant AI-based system on the diagnostic performance of radiologists in interpreting mammograms.

METHODS

We designed a multireader multicase study taking into account the evaluation of both interpretive and noninterpretive tasks. The study was approved by an institutional review board and is compliant with HIPAA. The dataset included 90 cancer-proven and 150 negative cases. The overall diagnostic performance was compared between the unaided vs aided reading condition. Intraclass correlation coefficient (ICC), Fleiss's kappa, and accuracy were used to quantify the agreement and performance on noninterpretive tasks. Reading time and perceived fatigue were used as comprehensive metrics to assess the efficiency of readers.

RESULTS

The average area under the receiver operating characteristic curve increased by 7.4% (95% CI, 4.5%-10%) with the concurrent assistance of the AI system (P <.001). On average, readers found 8% more cancers in the assisted reading condition. The ICC went from 0.6 (95% CI, 0.55-0.65) in the unassisted condition to 0.74 (95% CI, 0.70-0.78) for readings done with AI (P <.001). An overall decrease of 24% in reading time and a reduction in perceived fatigue was also found.

CONCLUSION

The incorporation of this AI system, capable of handling multiple image type, prior mammograms, and multiple outputs, improved the diagnostic proficiency of radiologists in identifying breast cancer while also reducing the time required for combined interpretive and noninterpretive tasks.

Performance of Digital Mammography-Based Artificial Intelligence Computer-Aided Diagnosis on Synthetic Mammography From Digital Breast Tomosynthesis

OBJECTIVE

To test the performance of an artificial intelligence-based computer-aided diagnosis (AI-CAD) designed for full-field digital mammography (FFDM) when applied to synthetic mammography (SM).

MATERIALS AND METHODS

We analyzed 501 women (mean age, 57 ± 11 years) who underwent preoperative mammography and breast cancer surgery. This cohort consisted of 1002 breasts, comprising 517 with cancer and 485 without. All patients underwent digital breast tomosynthesis (DBT) and FFDM during the preoperative workup. The SM is routinely reconstructed using DBT. Commercial AI-CAD (Lunit Insight MMG, version 1.1.7.2) was retrospectively applied to SM and FFDM to calculate the abnormality scores for each breast. The median abnormality scores were compared for the 517 breasts with cancer using the Wilcoxon signed-rank test. Calibration curves of abnormality scores were evaluated. The discrimination performance was analyzed using the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity using a 10% preset threshold. Sensitivity and specificity were further analyzed according to the mammographic and pathological characteristics. The results of SM and FFDM were compared.

RESULTS

AI-CAD demonstrated a significantly lower median abnormality score (71% vs. 96%, P < 0.001) and poorer calibration performance for SM than for FFDM. SM exhibited lower sensitivity (76.2% vs. 82.8%, P < 0.001), higher specificity (95.5% vs. 91.8%, P < 0.001), and comparable AUC (0.86 vs. 0.87, P = 0.127) than FFDM. SM showed lower sensitivity than FFDM in asymptomatic breasts, dense breasts, ductal carcinoma in situ, T1, N0, and hormone receptor-positive/human epidermal growth factor receptor 2-negative cancers but showed higher specificity in non-cancerous dense breasts.

CONCLUSION

AI-CAD showed lower abnormality scores and reduced calibration performance for SM than for FFDM. Furthermore, the 10% preset threshold resulted in different discrimination performances for the SM. Given these limitations, off-label application of the current AI-CAD to SM should be avoided.

Screening Mammography and Artificial Intelligence: A Comprehensive Systematic Review

Screening mammography is vital for early breast cancer detection, improving outcomes by identifying malignancies at treatable stages. Artificial intelligence has emerged as a tool to enhance diagnostic accuracy and reduce radiologists' workload in screening programs, though its full integration into clinical practice remains limited, necessitating a comprehensive review of its performance. This systematic review assesses artificial intelligence's effectiveness in screening mammography, focusing on diagnostic performance, reduction of false positives, and support for radiologists in clinical decision-making. A systematic search was conducted across PubMed, Embase, Web of Science, Cochrane Central, and Scopus for studies published between 2013 and 2024, including those evaluating artificial intelligence in mammography screening and reporting outcomes related to cancer detection, sensitivity, specificity, and workflow optimization. A total of 13 studies were analyzed, with data extracted on study characteristics, population demographics, artificial intelligence algorithms, and key outcomes. Artificial intelligence-assisted readings in screening mammography were found to be comparable or superior to traditional double readings by radiologists, reducing unnecessary recalls, improving specificity, and in some cases increasing cancer detection rates. Its integration into workflows showed potential for reducing radiologist workload while maintaining high diagnostic performance; however, challenges such as high false-positive rates and variations in artificial intelligence performance across patient subgroups remain concerns. Overall, artificial intelligence has the potential to enhance the efficiency and accuracy of breast cancer screening programs, and while it can reduce unnecessary recalls and alleviate radiologists' workloads, issues with false positives and demographic variations in accuracy highlight the need for further research. With ongoing refinement, artificial intelligence could become a valuable tool in routine mammography screening, augmenting radiologists' capabilities and improving patient care.

Validation of the Mirai model for predicting breast cancer risk in Mexican women

OBJECTIVES

To validate the performance of Mirai, a mammography-based deep learning model, in predicting breast cancer risk over a 1-5-year period in Mexican women.

METHODS

This retrospective single-center study included mammograms in Mexican women who underwent screening mammography between January 2014 and December 2016. For women with consecutive mammograms during the study period, only the initial mammogram was included. Pathology and imaging follow-up served as the reference standard. Model performance in the entire dataset was evaluated, including the concordance index (C-Index) and area under the receiver operating characteristic curve (AUC). Mirai's performance in terms of AUC was also evaluated between mammography systems (Hologic versus IMS). Clinical utility was evaluated by determining a cutoff point for Mirai's continuous risk index based on identifying the top 10% of patients in the high-risk category.

RESULTS

Of 3110 patients (median age 52.6 years ± 8.9), throughout the 5-year follow-up period, 3034 patients remained cancer-free, while 76 patients developed breast cancer. Mirai achieved a C-index of 0.63 (95% CI: 0.6-0.7) for the entire dataset. Mirai achieved a higher mean C-index in the Hologic subgroup (0.63 [95% CI: 0.5-0.7]) versus the IMS subgroup (0.55 [95% CI: 0.4-0.7]). With a Mirai index score > 0.029 (10% threshold) to identify high-risk individuals, the study revealed that individuals in the high-risk group had nearly three times the risk of developing breast cancer compared to those in the low-risk group.

CONCLUSIONS

Mirai has a moderate performance in predicting future breast cancer among Mexican women.

CRITICAL RELEVANCE STATEMENT

Prospective efforts should refine and apply the Mirai model, especially to minority populations and women aged between 30 and 40 years who are currently not targeted for routine screening.

KEY POINTS

The applicability of AI models to non-White, minority populations remains understudied. The Mirai model is linked to future cancer events in Mexican women. Further research is needed to enhance model performance and establish usage guidelines.

Artificial Intelligence (AI)-Based Computer-Assisted Detection and Diagnosis for Mammography: An Evidence-Based Review of Food and Drug Administration (FDA)-Cleared Tools for Screening Digital Breast Tomosynthesis (DBT)

In recent years, the emergence of new-generation deep learning-based artificial intelligence (AI) tools has reignited enthusiasm about the potential of computer-assisted detection (CADe) and diagnosis (CADx) for screening mammography. For screening mammography, digital breast tomosynthesis (DBT) combined with acquired digital 2D mammography or synthetic 2D mammography is widely used throughout the United States. As of this writing in July 2024, there are six Food and Drug Administration (FDA)-cleared AI-based CADe/x tools for DBT. These tools detect suspicious lesions on DBT and provide corresponding scores at the lesion and examination levels that reflect likelihood of malignancy. In this article, we review the evidence supporting the use of AI-based CADe/x for DBT. The published literature on this topic consists of multireader, multicase studies, retrospective analyses, and two "real-world" evaluations. These studies suggest that AI-based CADe/x could lead to improvements in sensitivity without compromising specificity and to improvements in efficiency. However, the overall published evidence is limited and includes only two small postimplementation clinical studies. Prospective studies and careful postimplementation clinical evaluation will be necessary to fully understand the impact of AI-based CADe/x on screening DBT outcomes.

Performance of artificial intelligence in 7533 consecutive prevalent screening mammograms from the BreastScreen Australia program

OBJECTIVES

To assess the performance of an artificial intelligence (AI) algorithm in the Australian mammography screening program which routinely uses two independent readers with arbitration of discordant results.

METHODS

A total of 7533 prevalent round mammograms from 2017 were available for analysis. The AI program classified mammograms into deciles on the basis of breast cancer (BC) risk. BC diagnoses, including invasive BC (IBC) and ductal carcinoma in situ (DCIS), included those from the prevalent round, interval cancers, and cancers identified in the subsequent screening round two years later. Performance was assessed by sensitivity, specificity, positive and negative predictive values, and the proportion of women recalled by the radiologists and identified as higher risk by AI.

RESULTS

Radiologists identified 54 women with IBC and 13 with DCIS with a recall rate of 9.7%. In contrast, 51 of 54 of the IBCs and 12/13 cases of DCIS were within the higher AI score group (score 10), a recall equivalent of 10.6% (a difference of 0.9% (CI -0.03 to 1.89%, p = 0.06). When IBCs were identified in the 2017 round, interval cancers classified as false negatives or with minimal signs in 2017, and cancers from the 2019 round were combined, the radiologists identified 54/67 and 59/67 were in the highest risk AI category (sensitivity 80.6% and 88.06 % respectively, a difference that was not different statistically).

CONCLUSIONS

As the performance of AI was comparable to that of expert radiologists, future AI roles in screening could include replacing one reader and supporting arbitration, reducing workload and false positive results.

CLINICAL RELEVANCE STATEMENT

AI analysis of consecutive prevalent screening mammograms from the Australian BreastScreen program demonstrated the algorithm's ability to match the cancer detection of experienced radiologists, additionally identifying five interval cancers (false negatives), and the majority of the false positive recalls.

KEY POINTS

• The AI program was almost as sensitive as the radiologists in terms of identifying prevalent lesions (51/54 for invasive breast cancer, 63/67 when including ductal carcinoma in situ). • If selected interval cancers and cancers identified in the subsequent screening round were included, the AI program identified more cancers than the radiologists (59/67 compared with 54/67, sensitivity 88.06 % and 80.6% respectively p = 0.24). • The high negative predictive value of a score of 1-9 would indicate a role for AI as a triage tool to reduce the recall rate (specifically false positives).

Barriers and facilitators of artificial intelligence conception and implementation for breast imaging diagnosis in clinical practice: a scoping review

OBJECTIVE

Although artificial intelligence (AI) has demonstrated promise in enhancing breast cancer diagnosis, the implementation of AI algorithms in clinical practice encounters various barriers. This scoping review aims to identify these barriers and facilitators to highlight key considerations for developing and implementing AI solutions in breast cancer imaging.

METHOD

A literature search was conducted from 2012 to 2022 in six databases (PubMed, Web of Science, CINHAL, Embase, IEEE, and ArXiv). The articles were included if some barriers and/or facilitators in the conception or implementation of AI in breast clinical imaging were described. We excluded research only focusing on performance, or with data not acquired in a clinical radiology setup and not involving real patients.

RESULTS

A total of 107 articles were included. We identified six major barriers related to data (B1), black box and trust (B2), algorithms and conception (B3), evaluation and validation (B4), legal, ethical, and economic issues (B5), and education (B6), and five major facilitators covering data (F1), clinical impact (F2), algorithms and conception (F3), evaluation and validation (F4), and education (F5).

CONCLUSION

This scoping review highlighted the need to carefully design, deploy, and evaluate AI solutions in clinical practice, involving all stakeholders to yield improvement in healthcare.

CLINICAL RELEVANCE STATEMENT

The identification of barriers and facilitators with suggested solutions can guide and inform future research, and stakeholders to improve the design and implementation of AI for breast cancer detection in clinical practice.

KEY POINTS

• Six major identified barriers were related to data; black-box and trust; algorithms and conception; evaluation and validation; legal, ethical, and economic issues; and education. • Five major identified facilitators were related to data, clinical impact, algorithms and conception, evaluation and validation, and education. • Coordinated implication of all stakeholders is required to improve breast cancer diagnosis with AI.

European validation of an image-derived AI-based short-term risk model for individualized breast cancer screening-a nested case-control study

Background

Image-derived artificial intelligence (AI)-based risk models for breast cancer have shown high discriminatory performances compared with clinical risk models based on family history and lifestyle factors. However, little is known about their generalizability across European screening settings. We therefore investigated the discriminatory performances of an AI-based risk model in European screening settings.

Methods

Using four European screening populations in three countries (Italy, Spain, Germany) screened between 2009 and 2020 for women aged 45-69, we performed a nested case-control study to assess the predictive performance of an AI-based risk model. In total, 739 women with incident breast cancers were included together with 7812 controls matched on year of study-entry. Mammographic features (density, microcalcifications, masses, left-right breast asymmetries of these features) were extracted using AI from negative digital mammograms at study-entry. Two-year absolute risks of breast cancer were predicted and assessed after two years of follow-up. Adjusted risk stratification performance metrics were reported per clinical guidelines.

Findings

The overall adjusted Area Under the receiver operating characteristic Curve (aAUC) of the AI risk model was 0.72 (95% CI 0.70-0.75) for breast cancers developed in four screening populations. In the 6.2% [529/8551] of women at high risk using the National Institute of Health and Care Excellence (NICE) guidelines thresholds, cancers were more likely diagnosed after 2 years follow-up, risk-ratio (RR) 6.7 (95% CI 5.6-8.0), compared with the 69% [5907/8551] of women classified at general risk by the model. Similar risk-ratios were observed across levels of mammographic density.

Interpretation

The AI risk model showed generalizable discriminatory performances across European populations and, predicted ∼30% of clinically relevant stage 2 and higher breast cancers in ∼6% of high-risk women who were sent home with a negative mammogram. Similar results were seen in women with fatty and dense breasts.

Funding

Swedish Research Council.

Comprehensive Overview of Computational Modeling and Artificial Intelligence in Pediatric Neurosurgery

In this chapter, we give an overview of artificial intelligence tools and their use thus far in pediatric neurosurgery. We discuss different machine learning algorithms from a data-driven approach in order to guide clinicians and scientists as they apply them to real-world datasets. We provide examples of their successful application as well as evaluate limitations and pitfalls specific to clinical use. Finally, we explore future directions and exciting new opportunities to take advantage of these tools as they continue to advance and evolve.

Artificial intelligence-based computer-assisted detection/diagnosis (AI-CAD) for screening mammography: Outcomes of AI-CAD in the mammographic interpretation workflow

Purpose

To evaluate the stand-alone diagnostic performances of AI-CAD and outcomes of AI-CAD detected abnormalities when applied to the mammographic interpretation workflow.

Methods

From January 2016 to December 2017, 6499 screening mammograms of 5228 women were collected from a single screening facility. Historic reads of three radiologists were used as radiologist interpretation. A commercially-available AI-CAD was used for analysis. One radiologist not involved in interpretation had retrospectively reviewed the abnormality features and assessed the significance (negligible vs. need recall) of the AI-CAD marks. Ground truth in terms of cancer, benign or absence of abnormality was confirmed according to histopathologic diagnosis or negative results on the next-round screen.

Results

Of the 6499 mammograms, 6282 (96.7%) were in the negative, 189 (2.9%) were in the benign, and 28 (0.4%) were in the cancer group. AI-CAD detected 5 (17.9%, 5 of 28) of the 9 cancers that were intially interpreted as negative. Of the 648 AI-CAD recalls, 89.0% (577 of 648) were marks seen on examinations in the negative group, and 267 (41.2%) of the AI-CAD marks were considered to be negligible. Stand-alone AI-CAD has significantly higher recall rates (10.0% vs. 3.4%, P < 0.001) with comparable sensitivity and cancer detection rates (P = 0.086 and 0.102, respectively) when compared to the radiologists' interpretation.

Conclusion

AI-CAD detected 17.9% additional cancers on screening mammography that were initially overlooked by the radiologists. In spite of the additional cancer detection, AI-CAD had significantly higher recall rates in the clinical workflow, in which 89.0% of AI-CAD marks are on negative mammograms.

New Approaches and Recommendations for Risk-Adapted Breast Cancer Screening

Population-based breast cancer screening using mammography as the gold standard imaging modality has been in clinical practice for over 40 years. However, the limitations of mammography in terms of sensitivity and high false-positive rates, particularly in high-risk women, challenge the indiscriminate nature of population-based screening. Additionally, in light of expanding research on new breast cancer risk factors, there is a growing consensus that breast cancer screening should move toward a risk-adapted approach. Recent advancements in breast imaging technology, including contrast material-enhanced mammography (CEM), ultrasound (US) (automated-breast US, Doppler, elastography US), and especially magnetic resonance imaging (MRI) (abbreviated, ultrafast, and contrast-agent free), may provide new opportunities for risk-adapted personalized screening strategies. Moreover, the integration of artificial intelligence and radiomics techniques has the potential to enhance the performance of risk-adapted screening. This review article summarizes the current evidence and challenges in breast cancer screening and highlights potential future perspectives for various imaging techniques in a risk-adapted breast cancer screening approach. EVIDENCE LEVEL: 1. TECHNICAL EFFICACY: Stage 5.

Quality assurance in Mammography: An overview

Since 1989, hundreds of thousands of lives have been saved worldwide by the widespread use of screening mammography alongside new developments in breast cancer treatment [1]. The ability of screening mammography to detect cancer early, when treatment is most effective, is optimized when it is performed in the highest quality manner and accessed by all eligible candidates. Currently, worldwide, there are over 14 guidance documents for mammographic quality [2]. Some countries, such as the United Kingdom (UK), monitor quality through a national screening program. In the United States (US), where 39 million mammograms are performed annually [3], there is not a national screening program, but the federal government mandates minimum quality standards for the performance of mammography. Among a consortium of European countries, the European Reference Organisation for Quality Assured Breast Screening and Diagnostic Services (EUREF) promotes voluntary adherence to European mammography quality standards. Setting quality standards at national or international levels ensures the uniformity of practice and identifies substandard practices in need of improvement, ultimately maximizing the benefit of screening mammography.

Expanding Horizons: The Realities of CAD, the Promise of Artificial Intelligence, and Machine Learning's Role in Breast Imaging beyond Screening Mammography

Artificial intelligence (AI) applications in mammography have gained significant popular attention; however, AI has the potential to revolutionize other aspects of breast imaging beyond simple lesion detection. AI has the potential to enhance risk assessment by combining conventional factors with imaging and improve lesion detection through a comparison with prior studies and considerations of symmetry. It also holds promise in ultrasound analysis and automated whole breast ultrasound, areas marked by unique challenges. AI's potential utility also extends to administrative tasks such as MQSA compliance, scheduling, and protocoling, which can reduce the radiologists' workload. However, adoption in breast imaging faces limitations in terms of data quality and standardization, generalizability, benchmarking performance, and integration into clinical workflows. Developing methods for radiologists to interpret AI decisions, and understanding patient perspectives to build trust in AI results, will be key future endeavors, with the ultimate aim of fostering more efficient radiology practices and better patient care.

Artificial Intelligence Applications in Breast Imaging: Current Status and Future Directions

Attempts to use computers to aid in the detection of breast malignancies date back more than 20 years. Despite significant interest and investment, this has historically led to minimal or no significant improvement in performance and outcomes with traditional computer-aided detection. However, recent advances in artificial intelligence and machine learning are now starting to deliver on the promise of improved performance. There are at present more than 20 FDA-approved AI applications for breast imaging, but adoption and utilization are widely variable and low overall. Breast imaging is unique and has aspects that create both opportunities and challenges for AI development and implementation. Breast cancer screening programs worldwide rely on screening mammography to reduce the morbidity and mortality of breast cancer, and many of the most exciting research projects and available AI applications focus on cancer detection for mammography. There are, however, multiple additional potential applications for AI in breast imaging, including decision support, risk assessment, breast density quantitation, workflow and triage, quality evaluation, response to neoadjuvant chemotherapy assessment, and image enhancement. In this review the current status, availability, and future directions of investigation of these applications are discussed, as well as the opportunities and barriers to more widespread utilization.

Automation Bias in Mammography: The Impact of Artificial Intelligence BI-RADS Suggestions on Reader Performance

Background Automation bias (the propensity for humans to favor suggestions from automated decision-making systems) is a known source of error in human-machine interactions, but its implications regarding artificial intelligence (AI)-aided mammography reading are unknown. Purpose To determine how automation bias can affect inexperienced, moderately experienced, and very experienced radiologists when reading mammograms with the aid of an artificial intelligence (AI) system. Materials and Methods In this prospective experiment, 27 radiologists read 50 mammograms and provided their Breast Imaging Reporting and Data System (BI-RADS) assessment assisted by a purported AI system. Mammograms were obtained between January 2017 and December 2019 and were presented in two randomized sets. The first was a training set of 10 mammograms, with the correct BI-RADS category suggested by the AI system. The second was a set of 40 mammograms in which an incorrect BI-RADS category was suggested for 12 mammograms. Reader performance, degree of bias in BI-RADS scoring, perceived accuracy of the AI system, and reader confidence in their own BI-RADS ratings were assessed using analysis of variance (ANOVA) and repeated-measures ANOVA followed by post hoc tests and Kruskal-Wallis tests followed by the Dunn post hoc test. Results The percentage of correctly rated mammograms by inexperienced (mean, 79.7% ± 11.7 [SD] vs 19.8% ± 14.0; P < .001; r = 0.93), moderately experienced (mean, 81.3% ± 10.1 vs 24.8% ± 11.6; P < .001; r = 0.96), and very experienced (mean, 82.3% ± 4.2 vs 45.5% ± 9.1; P = .003; r = 0.97) radiologists was significantly impacted by the correctness of the AI prediction of BI-RADS category. Inexperienced radiologists were significantly more likely to follow the suggestions of the purported AI when it incorrectly suggested a higher BI-RADS category than the actual ground truth compared with both moderately (mean degree of bias, 4.0 ± 1.8 vs 2.4 ± 1.5; P = .044; r = 0.46) and very (mean degree of bias, 4.0 ± 1.8 vs 1.2 ± 0.8; P = .009; r = 0.65) experienced readers. Conclusion The results show that inexperienced, moderately experienced, and very experienced radiologists reading mammograms are prone to automation bias when being supported by an AI-based system. This and other effects of human and machine interaction must be considered to ensure safe deployment and accurate diagnostic performance when combining human readers and AI. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Baltzer in this issue.

Beyond Breast Density: Risk Measures for Breast Cancer in Multiple Imaging Modalities

Breast density is an independent risk factor for breast cancer. In digital mammography and digital breast tomosynthesis, breast density is assessed visually using the four-category scale developed by the American College of Radiology Breast Imaging Reporting and Data System (5th edition as of November 2022). Epidemiologically based risk models, such as the Tyrer-Cuzick model (version 8), demonstrate superior modeling performance when mammographic density is incorporated. Beyond just density, a separate mammographic measure of breast cancer risk is parenchymal textural complexity. With advancements in radiomics and deep learning, mammographic textural patterns can be assessed quantitatively and incorporated into risk models. Other supplemental screening modalities, such as breast US and MRI, offer independent risk measures complementary to those derived from mammography. Breast US allows the two components of fibroglandular tissue (stromal and glandular) to be visualized separately in a manner that is not possible with mammography. A higher glandular component at screening breast US is associated with higher risk. With MRI, a higher background parenchymal enhancement of the fibroglandular tissue has also emerged as an imaging marker for risk assessment. Imaging markers observed at mammography, US, and MRI are powerful tools in refining breast cancer risk prediction, beyond mammographic density alone.

Artificial intelligence for digital breast tomosynthesis: Impact on diagnostic performance, reading times, and workload in the era of personalized screening

The ultimate goals of the application of artificial intelligence (AI) to digital breast tomosynthesis (DBT) are the reduction of reading times, the increase of diagnostic performance, and the reduction of interval cancer rates. In this review, after outlining the journey from computer-aided detection/diagnosis systems to AI applied to digital mammography (DM), we summarize the results of studies where AI was applied to DBT, noting that long-term advantages of DBT screening and its crucial ability to decrease the interval cancer rate are still under scrutiny. AI has shown the capability to overcome some shortcomings of DBT in the screening setting by improving diagnostic performance and by reducing recall rates (from -2 % to -27 %) and reading times (up to -53 %, with an average 20 % reduction), but the ability of AI to reduce interval cancer rates has not yet been clearly investigated. Prospective validation is needed to assess the cost-effectiveness and real-world impact of AI models assisting DBT interpretation, especially in large-scale studies with low breast cancer prevalence. Finally, we focus on the incoming era of personalized and risk-stratified screening that will first see the application of contrast-enhanced breast imaging to screen women with extremely dense breasts. As the diagnostic advantage of DBT over DM was concentrated in this category, we try to understand if the application of AI to DM in the remaining cohorts of women with heterogeneously dense or non-dense breast could close the gap in diagnostic performance between DM and DBT, thus neutralizing the usefulness of AI application to DBT.

Artificial Intelligence in Breast Cancer Screening: Evaluation of FDA Device Regulation and Future Recommendations

Importance

Contemporary approaches to artificial intelligence (AI) based on deep learning have generated interest in the application of AI to breast cancer screening (BCS). The US Food and Drug Administration (FDA) has approved several next-generation AI products indicated for BCS in recent years; however, questions regarding their accuracy, appropriate use, and clinical utility remain.

Objectives

To describe the current FDA regulatory process for AI products, summarize the evidence used to support FDA clearance and approval of AI products indicated for BCS, consider the advantages and limitations of current regulatory approaches, and suggest ways to improve the current system.

Evidence Review

Premarket notifications and other publicly available documents used for FDA clearance and approval of AI products indicated for BCS from January 1, 2017, to December 31, 2021.

Findings

Nine AI products indicated for BCS for identification of suggestive lesions and mammogram triage were included. Most of the products had been cleared through the 510(k) pathway, and all clearances were based on previously collected retrospective data; 6 products used multicenter designs; 7 products used enriched data; and 4 lacked details on whether products were externally validated. Test performance measures, including sensitivity, specificity, and area under the curve, were the main outcomes reported. Most of the devices used tissue biopsy as the criterion standard for BCS accuracy evaluation. Other clinical outcome measures, including cancer stage at diagnosis and interval cancer detection, were not reported for any of the devices.

Conclusions and Relevance

The findings of this review suggest important gaps in reporting of data sources, data set type, validation approach, and clinical utility assessment. As AI-assisted reading becomes more widespread in BCS and other radiologic examinations, strengthened FDA evidentiary regulatory standards, development of postmarketing surveillance, a focus on clinically meaningful outcomes, and stakeholder engagement will be critical for ensuring the safety and efficacy of these products.

Artificial Intelligence in Clinical Practice: Implementation Considerations and Barriers

The rapid growth of artificial intelligence (AI) in radiology has led to Food and Drug Administration clearance of more than 20 AI algorithms for breast imaging. The steps involved in the clinical implementation of an AI product include identifying all stakeholders, selecting the appropriate product to purchase, evaluating it with a local data set, integrating it into the workflow, and monitoring its performance over time. Despite the potential benefits of improved quality and increased efficiency with AI, several barriers, such as high costs and liability concerns, may limit its widespread implementation. This article lists currently available AI products for breast imaging, describes the key elements of clinical implementation, and discusses barriers to clinical implementation.

External Validation of a Mammography-Derived AI-Based Risk Model in a U.S. Breast Cancer Screening Cohort of White and Black Women

Despite the demonstrated potential of artificial intelligence (AI) in breast cancer risk assessment for personalizing screening recommendations, further validation is required regarding AI model bias and generalizability. We performed external validation on a U.S. screening cohort of a mammography-derived AI breast cancer risk model originally developed for European screening cohorts. We retrospectively identified 176 breast cancers with exams 3 months to 2 years prior to cancer diagnosis and a random sample of 4963 controls from women with at least one-year negative follow-up. A risk score for each woman was calculated via the AI risk model. Age-adjusted areas under the ROC curves (AUCs) were estimated for the entire cohort and separately for White and Black women. The Gail 5-year risk model was also evaluated for comparison. The overall AUC was 0.68 (95% CIs 0.64−0.72) for all women, 0.67 (0.61−0.72) for White women, and 0.70 (0.65−0.76) for Black women. The AI risk model significantly outperformed the Gail risk model for all women p < 0.01 and for Black women p < 0.01, but not for White women p = 0.38. The performance of the mammography-derived AI risk model was comparable to previously reported European validation results; non-significantly different when comparing White and Black women; and overall, significantly higher than that of the Gail model.

Artificial intelligence in mammographic phenotyping of breast cancer risk: a narrative review

BACKGROUND

Improved breast cancer risk assessment models are needed to enable personalized screening strategies that achieve better harm-to-benefit ratio based on earlier detection and better breast cancer outcomes than existing screening guidelines. Computational mammographic phenotypes have demonstrated a promising role in breast cancer risk prediction. With the recent exponential growth of computational efficiency, the artificial intelligence (AI) revolution, driven by the introduction of deep learning, has expanded the utility of imaging in predictive models. Consequently, AI-based imaging-derived data has led to some of the most promising tools for precision breast cancer screening.

MAIN BODY

This review aims to synthesize the current state-of-the-art applications of AI in mammographic phenotyping of breast cancer risk. We discuss the fundamentals of AI and explore the computing advancements that have made AI-based image analysis essential in refining breast cancer risk assessment. Specifically, we discuss the use of data derived from digital mammography as well as digital breast tomosynthesis. Different aspects of breast cancer risk assessment are targeted including (a) robust and reproducible evaluations of breast density, a well-established breast cancer risk factor, (b) assessment of a woman's inherent breast cancer risk, and (c) identification of women who are likely to be diagnosed with breast cancers after a negative or routine screen due to masking or the rapid and aggressive growth of a tumor. Lastly, we discuss AI challenges unique to the computational analysis of mammographic imaging as well as future directions for this promising research field.

CONCLUSIONS

We provide a useful reference for AI researchers investigating image-based breast cancer risk assessment while indicating key priorities and challenges that, if properly addressed, could accelerate the implementation of AI-assisted risk stratification to future refine and individualize breast cancer screening strategies.