A risk model for digital breast tomosynthesis to predict breast cancer and guide clinical care

Wasm-iCARE: a portable and privacy-preserving web module to build, validate, and apply absolute risk models

Objectives

Absolute risk models estimate an individual's future disease risk over a specified time interval. Applications utilizing server-side risk tooling, the R-based iCARE (R-iCARE), to build, validate, and apply absolute risk models, face limitations in portability and privacy due to their need for circulating user data in remote servers for operation. We overcome this by porting iCARE to the web platform.

Materials and Methods

We refactored R-iCARE into a Python package (Py-iCARE) and then compiled it to WebAssembly (Wasm-iCARE)-a portable web module, which operates within the privacy of the user's device.

Results

We showcase the portability and privacy of Wasm-iCARE through 2 applications: for researchers to statistically validate risk models and to deliver them to end-users. Both applications run entirely on the client side, requiring no downloads or installations, and keep user data on-device during risk calculation.

Conclusions

Wasm-iCARE fosters accessible and privacy-preserving risk tools, accelerating their validation and delivery.

Artificial Intelligence-Powered Imaging Biomarker Based on Mammography for Breast Cancer Risk Prediction

The purposes of this study were to develop an artificial intelligence (AI) model for future breast cancer risk prediction based on mammographic images, investigate the feasibility of the AI model, and compare the AI model, clinical statistical risk models, and Mirai, a state of-the art deep learning algorithm based on screening mammograms for 1-5-year breast cancer risk prediction. We trained and developed a deep learning model using a total of 36,995 serial mammographic examinations from 21,438 women (cancer-enriched mammograms, 17.5%). To determine the feasibility of the AI prediction model, mammograms and detailed clinical information were collected. C-indices and area under the receiver operating characteristic curves (AUCs) for 1-5-year outcomes were obtained. We compared the AUCs of our AI prediction model, Mirai, and clinical statistical risk models, including the Tyrer-Cuzick (TC) model and Gail model, using DeLong's test. A total of 16,894 mammograms were independently collected for external validation, of which 4002 were followed by a cancer diagnosis within 5 years. Our AI prediction model obtained a C-index of 0.76, with AUCs of 0.90, 0.84, 0.81, 0.78, and 0.81, to predict the 1-5-year risks. Our AI prediction model showed significantly higher AUCs than those of the TC model (AUC: 0.57; p < 0.001) and Gail model (AUC: 0.52; p < 0.001), and achieved similar performance to Mirai. The deep learning AI model using mammograms and AI-powered imaging biomarkers has substantial potential to advance accurate breast cancer risk prediction.

Breast cancer risk prediction using machine learning: a systematic review

Background

Breast cancer is the leading cause of cancer-related fatalities among women worldwide. Conventional screening and risk prediction models primarily rely on demographic and patient clinical history to devise policies and estimate likelihood. However, recent advancements in artificial intelligence (AI) techniques, particularly deep learning (DL), have shown promise in the development of personalized risk models. These models leverage individual patient information obtained from medical imaging and associated reports. In this systematic review, we thoroughly investigated the existing literature on the application of DL to digital mammography, radiomics, genomics, and clinical information for breast cancer risk assessment. We critically analyzed these studies and discussed their findings, highlighting the promising prospects of DL techniques for breast cancer risk prediction. Additionally, we explored ongoing research initiatives and potential future applications of AI-driven approaches to further improve breast cancer risk prediction, thereby facilitating more effective screening and personalized risk management strategies.

Objective and methods

This study presents a comprehensive overview of imaging and non-imaging features used in breast cancer risk prediction using traditional and AI models. The features reviewed in this study included imaging, radiomics, genomics, and clinical features. Furthermore, this survey systematically presented DL methods developed for breast cancer risk prediction, aiming to be useful for both beginners and advanced-level researchers.

Results

A total of 600 articles were identified, 20 of which met the set criteria and were selected. Parallel benchmarking of DL models, along with natural language processing (NLP) applied to imaging and non-imaging features, could allow clinicians and researchers to gain greater awareness as they consider the clinical deployment or development of new models. This review provides a comprehensive guide for understanding the current status of breast cancer risk assessment using AI.

Conclusion

This study offers investigators a different perspective on the use of AI for breast cancer risk prediction, incorporating numerous imaging and non-imaging features.

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.

Implementing the National Dense Breast Reporting Standard, Expanding Supplemental Screening Using Current Guidelines, and the Proposed Find It Early Act

Thirty-eight states and the District of Columbia (DC) have dense breast notification laws that mandate varying levels of patient notification about breast density after a mammogram, and these cover over 90% of American women. On March 10, 2023, the Food and Drug Administration issued a final rule amending regulations under the Mammography Quality Standards Act for a national dense breast reporting standard for both patient results letters and mammogram reports. Effective September 10, 2024, letters will be required to tell a woman her breasts are "dense" or "not dense," that dense tissue makes it harder to find cancers on a mammogram, and that it increases the risk of developing cancer. Women with dense breasts will also be told that other imaging tests in addition to a mammogram may help find cancers. The specific density category can be added (eg, if mandated by a state "inform" law). Reports to providers must include the Breast Imaging Reporting and Data System density category. Implementing appropriate supplemental screening should be based on patient risk for missed breast cancer on mammography; such assessment should include consideration of breast density and other risk factors. This article discusses strategies for implementation. Currently 21 states and DC have varying insurance laws for supplemental breast imaging; in addition, Oklahoma requires coverage for diagnostic breast imaging. A federal insurance bill, the Find It Early Act, has been introduced that would ensure no-cost screening and diagnostic imaging for women with dense breasts or at increased risk and close loopholes in state laws.

Breast density analysis of digital breast tomosynthesis

Mammography shifted to digital breast tomosynthesis (DBT) in the US. An automated percentage of breast density (PD) technique designed for two-dimensional (2D) applications was evaluated with DBT using several breast cancer risk prediction measures: normalized-volumetric; dense volume; applied to the volume slices and averaged (slice-mean); and applied to synthetic 2D images. Volumetric measures were derived theoretically. PD was modeled as a function of compressed breast thickness (CBT). The mean and standard deviation of the pixel values were investigated. A matched case-control (CC) study (n = 426 pairs) was evaluated. Odd ratios (ORs) were estimated with 95% confidence intervals. ORs were significant for PD: identical for volumetric and slice-mean measures [OR = 1.43 (1.18, 1.72)] and [OR = 1.44 (1.18, 1.75)] for synthetic images. A 2nd degree polynomial (concave-down) was used to model PD as a function of CBT: location of the maximum PD value was similar across CCs, occurring at 0.41 × CBT, and PD was significant [OR = 1.47 (1.21, 1.78)]. The means from the volume and synthetic images were also significant [ORs ~ 1.31 (1.09, 1.57)]. An alternative standardized 2D synthetic image was constructed, where each pixel value represents the percentage of breast density above its location. Several measures were significant and an alternative method for constructing a standardized 2D synthetic image was produced.

Breast Cancer Risk Reduction: Current Status and Emerging Trends to Increase Efficacy and Reduce Toxicity of Preventive Medication

The primary prevention of breast cancer is a worthwhile goal for which the efficacy of antiestrogens is well established. However, implementation has been problematic related to low prioritization by providers and the reluctance of high-risk women to experience medication side effects. Emerging solutions include improved risk estimation through the use of polygenic risk scores and the application of radiomics to screening mammograms; and optimization of medication dose to limit toxicity. The identification of agents to prevent estrogen receptor negative or HER2-positive tumors is being pursued, but personalization of medical risk reduction requires the prediction of tumor subtypes.

Combining Molecular and Radiomic Features for Risk Assessment in Breast Cancer

Breast cancer risk is highly variable within the population and current research is leading the shift toward personalized medicine. By accurately assessing an individual woman's risk, we can reduce the risk of over/undertreatment by preventing unnecessary procedures or by elevating screening procedures. Breast density measured from conventional mammography has been established as one of the most dominant risk factors for breast cancer; however, it is currently limited by its ability to characterize more complex breast parenchymal patterns that have been shown to provide additional information to strengthen cancer risk models. Molecular factors ranging from high penetrance, or high likelihood that a mutation will show signs and symptoms of the disease, to combinations of gene mutations with low penetrance have shown promise for augmenting risk assessment. Although imaging biomarkers and molecular biomarkers have both individually demonstrated improved performance in risk assessment, few studies have evaluated them together. This review aims to highlight the current state of the art in breast cancer risk assessment using imaging and genetic biomarkers.

A Clinical Risk Model for Personalized Screening and Prevention of Breast Cancer

BACKGROUND

Image-derived artificial intelligence (AI) risk models have shown promise in identifying high-risk women in the short term. The long-term performance of image-derived risk models expanded with clinical factors has not been investigated.

METHODS

We performed a case-cohort study of 8110 women aged 40-74 randomly selected from a Swedish mammography screening cohort initiated in 2010 together with 1661 incident BCs diagnosed before January 2022. The imaging-only AI risk model extracted mammographic features and age at screening. Additional lifestyle/familial risk factors were incorporated into the lifestyle/familial-expanded AI model. Absolute risks were calculated using the two models and the clinical Tyrer-Cuzick v8 model. Age-adjusted model performances were compared across the 10-year follow-up.

RESULTS

The AUCs of the lifestyle/familial-expanded AI risk model ranged from 0.75 (95%CI: 0.70-0.80) to 0.68 (95%CI: 0.66-0.69) 1-10 years after study entry. Corresponding AUCs were 0.72 (95%CI: 0.66-0.78) to 0.65 (95%CI: 0.63-0.66) for the imaging-only model and 0.62 (95%CI: 0.55-0.68) to 0.60 (95%CI: 0.58-0.61) for Tyrer-Cuzick v8. The increased performances were observed in multiple risk subgroups and cancer subtypes. Among the 5% of women at highest risk, the PPV was 5.8% using the lifestyle/familial-expanded model compared with 5.3% using the imaging-only model, p < 0.01, and 4.6% for Tyrer-Cuzick, p < 0.01.

CONCLUSIONS

The lifestyle/familial-expanded AI risk model showed higher performance for both long-term and short-term risk assessment compared with imaging-only and Tyrer-Cuzick models.

Long-Term Performance of an Image-Based Short-Term Risk Model for Breast Cancer

PURPOSE

Image-derived artificial intelligence-based short-term risk models for breast cancer have shown high discriminatory performance compared with traditional lifestyle/familial-based risk models. The long-term performance of image-derived risk models has not been investigated.

METHODS

We performed a case-cohort study of 8,604 randomly selected women within a mammography screening cohort initiated in 2010 in Sweden for women age 40-74 years. Mammograms, age, lifestyle, and familial risk factors were collected at study entry. In all, 2,028 incident breast cancers were identified through register matching in May 2022 (206 incident breast cancers were found in the subcohort). The image-based model extracted mammographic features (density, microcalcifications, masses, and left-right breast asymmetries of these features) and age from study entry mammograms. The Tyrer-Cuzick v8 risk model incorporates self-reported lifestyle and familial risk factors and mammographic density to estimate risk. Absolute risks were estimated, and age-adjusted AUC model performances (aAUCs) were compared across the 10-year period.

RESULTS

The aAUCs of the image-based risk model ranged from 0.74 (95% CI, 0.70 to 0.78) to 0.65 (95% CI, 0.63 to 0.66) for breast cancers developed 1-10 years after study entry; the corresponding Tyrer-Cuzick aAUCs were 0.62 (95% CI, 0.56 to 0.67) to 0.60 (95% CI, 0.58 to 0.61). For symptomatic cancers, the aAUCs for the image-based model were ≥0.75 during the first 3 years. Women with high and low mammographic density showed similar aAUCs. Throughout the 10-year follow-up, 20% of all women with breast cancers were deemed high-risk at study entry by the image-based risk model compared with 7.1% using the lifestyle familial-based model (P < .01).

CONCLUSION

The image-based risk model outperformed the Tyrer-Cuzick v8 model for both short-term and long-term risk assessment and could be used to identify women who may benefit from supplemental screening and risk reduction strategies.

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.

Breast Density Analysis Using Digital Breast Tomosynthesis

We evaluated an automated percentage of breast density (BD) technique (PDa) with digital breast tomosynthesis (DBT) data. The approach is based on the wavelet expansion followed by analyzing signal dependent noise. Several measures were investigated as risk factors: normalized volumetric; total dense volume; average of the DBT slices (slice-mean); a two-dimensional (2D) metric applied to the synthetic images; and the mean and standard deviations of the pixel values. Volumetric measures were derived theoretically, and PDa was modeled as a function of compressed breast thickness. An alternative method for constructing synthetic 2D mammograms was investigated using the volume results. A matched case-control study (n = 426 pairs) was analyzed. Conditional logistic regression modeling, controlling body mass index and ethnicity, was used to estimate odds ratios (ORs) for each measure with 95% confidence intervals provided parenthetically. There were several significant findings: volumetric measure [OR = 1.43 (1.18, 1.72)], which produced an identical OR as the slice-mean measure as predicted; [OR =1.44 (1.18, 1.75)] when applied to the synthetic images; and mean of the pixel values (volume or 2D synthetic) [ORs ~ 1.31 (1.09, 1.57)]. PDa was modeled as 2nd degree polynomial (concave-down): its maximum value occurred at 0.41×(compressed breast thickness), which was similar across case-control groups, and was significant from this position [OR = 1.47 (1.21, 1.78)]. A standardized 2D synthetic image was produced, where each pixel value represents the percentage of BD above its location. The significant findings indicate the validity of the technique. Derivations supported by empirical analyses produced a new synthetic 2D standardized image technique. Ancillary to the objectives, the results provide evidence for understanding the percentage of BD measure applied to 2D mammograms. Notwithstanding the findings, the study design provides a template for investigating other measures such as texture.

Breast cancer screening with digital breast tomosynthesis: comparison of different reading strategies implementing artificial intelligence

Objectives

Digital 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.

Methods

An 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.

Results

By 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).

Conclusion

AI 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.

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.