Individual and combined effects of age, breast density, and hormone replacement therapy use on the accuracy of screening mammography

Breast Cancer Screening With Automated Breast US and Mammography vs Handheld US and Mammography in Women With Dense Breasts in a Real-World Clinical Setting

OBJECTIVE

We compared the performance of 2 breast cancer screening approaches, automated breast US (ABUS) with same-day mammography (ABUS/MG) and handheld US (HHUS) with same-day mammography (HHUS/MG), in women with dense breasts to better understand the relative usefulness of ABUS and HHUS in a real-world clinical setting.

METHODS

In this institutional review board-approved, retrospective observational study, we evaluated all ABUS/MG and HHUS/MG screening examinations performed at our institution from May 2013 to September 2021. BI-RADS categories, biopsy pathology results, and diagnostic test characteristics (eg, sensitivity, specificity) were compared between the 2 screening approaches using Fisher's exact test.

RESULTS

A total of 1120 women with dense breasts were included in this study, with 852 undergoing ABUS/MG and 268 undergoing HHUS/MG. The sensitivities of ABUS/MG and HHUS/MG were 100% (5/5) and 75.0% (3/4), respectively, which was not a statistically significant difference (P  = .444). The ABUS/MG approach demonstrated a slightly higher specificity (97.4% [825/847] vs 94.3% [249/264]; P = .028), higher accuracy (97.4% [830/852] vs 94.0% [252/268]; P = .011), and lower biopsy recommendation rate (3.2% [27/852] vs 6.7% [18/268]; P = .019) than the HHUS/MG approach in our patient population.

CONCLUSION

Our findings suggest that ABUS/MG performs comparably with HHUS/MG as a breast cancer screening approach in women with dense breasts in a real-world clinical setting, with the ABUS/MG approach demonstrating a similar sensitivity and slightly higher specificity than the HHUS/MG approach. Additional variables, such as patient experience and physician time, may help determine which imaging approach to employ in specific clinical settings.

Hormone therapy in women with benign breast disease - What little is known and suggestions for clinical implementation

Benign breast disease encompasses a spectrum of lesions within the breast. While some lesions pose no increase in risk, others may elevate the likelihood of developing breast cancer by four- to five-fold. This necessitates a personalized approach to screening and lifestyle optimization for women. The menopausal transition is a critical time for the development of benign breast lesions. Increased detection can be attributed to the heightened precision and utilization of screening mammography, with or without the use of supplemental imaging. While it is widely acknowledged that combined hormone therapy involving estrogen and progesterone may elevate the risk of breast cancer, data from the Women's Health Initiative (WHI) indicates that estrogen-alone therapies may actually reduce the overall risk of cancer. Despite this general understanding, there is a notable gap in information regarding the impact of hormone therapy on the risk profile of women with specific benign breast lesions. This review comprehensively examines various benign breast lesions, delving into their pathophysiology and management. The goal is to enhance our understanding of when and how to judiciously prescribe hormone therapy, particularly in the context of specific benign breast conditions. By bridging this knowledge gap, the review provides valuable insights into optimizing healthcare strategies for women with benign breast disease, and offers a foundation for more informed decision-making regarding hormone therapy.

Menopausal hormone therapy and breast cancer risk: a population-based cohort study of 1.3 million women in Norway

BACKGROUND

It is important to monitor the association between menopausal hormone therapy (HT) use and breast cancer (BC) risk with contemporary estimates, and specifically focus on HT types and new drugs.

METHODS

We estimated hazard ratios (HR) of BC risk according to HT type, administration route and individual drugs, overall and stratified by body mass index (BMI), molecular subtype and detection mode, with non-HT use as reference.

RESULTS

We included 1,275,783 women, 45+ years, followed from 2004, for a median of 12.7 years. Oral oestrogen combined with daily progestin was associated with the highest risk of BC (HR 2.42, 95% confidence interval (CI) 2.31-2.54), with drug-specific HRs ranging from Cliovelle®: 1.63 (95% CI 1.35-1.96) to Kliogest®: 2.67 (2.37-3.00). Vaginal oestradiol was not associated with BC risk. HT use was more strongly associated with luminal A cancer (HR 1.97, 95% CI 1.86-2.09) than other molecular subtypes, and more strongly with interval (HR 2.00, 95% CI: 1.83-2.30) than screen-detected (HR 1.33, 95% CI 1.26-1.41) BC in women 50-71 years. HRs for HT use decreased with increasing BMI.

CONCLUSIONS

The use of oral and transdermal HT was associated with an increased risk of BC. The associations varied according to HT type, individual drugs, molecular subtype, detection mode and BMI.

Interval Cancers in Understanding Screening Outcomes

Interval breast cancers are not detected at routine screening and are diagnosed in the interval between screening examinations. A variety of factors contribute to interval cancers, including patient and tumor characteristics as well as the screening technique and frequency. The interval cancer rate is an important metric by which the effectiveness of screening may be assessed and may serve as a surrogate for mortality benefit.

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.

Test-set results can predict participants' development in breast-screen cancer detection: An observational cohort study

Background and Aim

Test-sets are standardized assessments used to evaluate reader performance in breast screening. Understanding how test-set results affect real-world performance can help refine their use as a quality improvement tool. The aim of this study is to explore if mammographic test-set results could identify breast-screening readers who improved their cancer detection in association with test-set training.

Methods

Test-set results of 41 participants were linked to their annual cancer detection rate change in two periods oriented around their first test-set participation year. Correlation tests and a multiple linear regression model investigated the relationship between each metric in the test-set results and the change in detection rates. Additionally, participants were divided based on their improvement status between the two periods, and Mann-Whitney U test was used to determine if the subgroups differed in their test-set metrics.

Results

Test-set records indicated multiple significant correlations with the change in breast cancer detection rate: a moderate positive correlation with sensitivity (0.688, p < 0.001), a moderate negative correlation with specificity (-0.528, p < 0.001), and a low to moderate positive correlation with lesion sensitivity (0.469, p = 0.002), and the number of years screen-reading mammograms (0.365, p = 0.02). In addition, the overall regression was statistically significant (F (2,38) = 18.456 p < 0.001), with an R² of 0.493 (adjusted R² = 0.466) based on sensitivity (F = 27.132, p < 0.001) and specificity (F = 9.78, p = 0.003). Subgrouping the cohort based on the change in cancer detection indicated that the improved group is significantly higher in sensitivity (p < 0.001) and lesion sensitivity (p = 0.02) but lower in specificity (p = 0.003).

Conclusion

Sensitivity and specificity are the strongest test-set performance measures to predict the change in breast cancer detection in real-world breast screening settings following test-set participation.

The distribution of breast density in women aged 18 years and older

PURPOSE

Age and body mass index (BMI) are critical considerations when assessing individual breast cancer risk, particularly for women with dense breasts. However, age- and BMI-standardized estimates of breast density are not available for screen-aged women, and little is known about the distribution of breast density in women aged < 40. This cross-sectional study uses three different modalities: optical breast spectroscopy (OBS), dual-energy X-ray absorptiometry (DXA), and mammography, to describe the distributions of breast density across categories of age and BMI.

METHODS

Breast density measures were estimated for 1,961 Australian women aged 18-97 years using OBS (%water and %water + %collagen). Of these, 935 women had DXA measures (percent and absolute fibroglandular dense volume, %FGV and FGV, respectively) and 354 had conventional mammographic measures (percent and absolute dense area). The distributions for each breast density measure were described across categories of age and BMI.

RESULTS

The mean age was 38 years (standard deviation = 15). Median breast density measures decreased with age and BMI for all three modalities, except for DXA-FGV, which increased with BMI and decreased after age 30. The variation in breast density measures was largest for younger women and decreased with increasing age and BMI.

CONCLUSION

This unique study describes the distribution of breast density measures for women aged 18-97 using alternative and conventional modalities of measurement. While this study is the largest of its kind, larger sample sizes are needed to provide clinically useful age-standardized measures to identify women with high breast density for their age or BMI.

Patient Experience of Women With Dense Breasts Undergoing Screening Contrast-Enhanced Mammography

OBJECTIVE

We investigated patient experience with screening contrast-enhanced mammography (CEM) to determine whether a general population of women with dense breasts would accept CEM in a screening setting.

METHODS

In this institutional review board-approved prospective study, patients with heterogeneous and extremely dense breasts on their mammogram were invited to undergo screening CEM and complete pre-CEM and post-CEM surveys. On the pre-CEM survey, patients were asked about their attitudes regarding supplemental screening in general. On the post-CEM survey, patients were asked about their experience undergoing screening CEM, including causes and severity of any discomfort and whether they would consider undergoing screening CEM again in the future or recommend it to a friend.

RESULTS

One hundred sixty-three women were surveyed before and after screening CEM. Most patients, 97.5% (159/163), reported minimal or no unpleasantness associated with undergoing screening CEM. In addition, 91.4% (149/163) said they would probably or very likely undergo screening CEM in the future if it cost the same as a traditional screening mammogram, and 95.1% (155/163) said they would probably or very likely recommend screening CEM to a friend. Patients in this study, who were all willing to undergo CEM, more frequently reported a family history of breast cancer than a comparison cohort of women with dense breasts (58.2% vs 47.1%, P = .027).

CONCLUSION

Patients from a general population of women with dense breasts reported a positive experience undergoing screening CEM, suggesting screening CEM might be well received by this patient population, particularly if the cost was comparable with traditional screening mammography.

Impact of AI for Digital Breast Tomosynthesis on Breast Cancer Detection and Interpretation Time

Purpose To develop an artificial intelligence (AI) model for the diagnosis of breast cancer on digital breast tomosynthesis (DBT) images and to investigate whether it could improve diagnostic accuracy and reduce radiologist reading time. Materials and Methods A deep learning AI algorithm was developed and validated for DBT with retrospectively collected examinations (January 2010 to December 2021) from 14 institutions in the United States and South Korea. A multicenter reader study was performed to compare the performance of 15 radiologists (seven breast specialists, eight general radiologists) in interpreting DBT examinations in 258 women (mean age, 56 years ± 13.41 [SD]), including 65 cancer cases, with and without the use of AI. Area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and reading time were evaluated. Results The AUC for stand-alone AI performance was 0.93 (95% CI: 0.92, 0.94). With AI, radiologists' AUC improved from 0.90 (95% CI: 0.86, 0.93) to 0.92 (95% CI: 0.88, 0.96) (P = .003) in the reader study. AI showed higher specificity (89.64% [95% CI: 85.34%, 93.94%]) than radiologists (77.34% [95% CI: 75.82%, 78.87%]) (P < .001). When reading with AI, radiologists' sensitivity increased from 85.44% (95% CI: 83.22%, 87.65%) to 87.69% (95% CI: 85.63%, 89.75%) (P = .04), with no evidence of a difference in specificity. Reading time decreased from 54.41 seconds (95% CI: 52.56, 56.27) without AI to 48.52 seconds (95% CI: 46.79, 50.25) with AI (P < .001). Interreader agreement measured by Fleiss κ increased from 0.59 to 0.62. Conclusion The AI model showed better diagnostic accuracy than radiologists in breast cancer detection, as well as reduced reading times. The concurrent use of AI in DBT interpretation could improve both accuracy and efficiency. Keywords: Breast, Computer-Aided Diagnosis (CAD), Tomosynthesis, Artificial Intelligence, Digital Breast Tomosynthesis, Breast Cancer, Computer-Aided Detection, Screening Supplemental material is available for this article. © RSNA, 2024 See also the commentary by Bae in this issue.

Association of breast cancer with quantitative mammographic density measures for women receiving contrast-enhanced mammography

Women with high mammographic density have an increased risk of breast cancer. They may be offered contrast-enhanced mammography to improve breast cancer screening performance. Using a cohort of women receiving contrast-enhanced mammography, we evaluated whether conventional and modified mammographic density measures were associated with breast cancer. Sixty-six patients with newly diagnosed unilateral breast cancer were frequency matched on the basis of age to 133 cancer-free control individuals. On low-energy craniocaudal contrast-enhanced mammograms (equivalent to standard mammograms), we measured quantitative mammographic density using CUMULUS software at the conventional intensity threshold ("Cumulus") and higher-than-conventional thresholds ("Altocumulus," "Cirrocumulus"). The measures were standardized to enable estimation of odds ratio per adjusted standard deviation (OPERA). In multivariable logistic regression of case-control status, only the highest-intensity measure (Cirrocumulus) was statistically significantly associated with breast cancer (OPERA = 1.40, 95% confidence interval = 1.04 to 1.89). Conventional Cumulus did not contribute to model fit. For women receiving contrast-enhanced mammography, Cirrocumulus mammographic density may better predict breast cancer than conventional quantitative mammographic density.

The impact of breast density notification on psychosocial outcomes in racial and ethnic minorities: A systematic review

BACKGROUND

High breast density is an independent risk factor for breast cancer and decreases the sensitivity of mammography. This systematic review synthesizes the evidence on the impact of breast density (BD) information and/or notification on women's psychosocial outcomes among women from racial and ethnic minority groups.

METHODS

A systematic search was performed in March 2023, and the articles were identified using CINHAL, Embase, Medline, and PsychInfo databases. The search strategy combined the terms "breast", "density", "notification" and synonyms. The authors specifically kept the search terms broad and did not include terms related to race and ethnicity. Full-text articles were reviewed for analysis by race, ethnicity and primary language of participants. Two authors evaluated the eligibility of studies with verification from the study team, extracted and crosschecked data, and assessed the risk of bias.

RESULTS

Of 1784 articles, 32 articles published from 2003 to 2023 were included. Thirty-one studies were conducted in the United States and one in Australia, with 28 quantitative and four qualitative methodologies. The overall results in terms of breast density awareness, knowledge, communication with healthcare professionals, screening intentions and supplemental screening practice were heterogenous across studies. Barriers to understanding BD notifications and intentions/access to supplemental screening among racial and ethnic minorities included socioeconomic factors, language, health literacy and medical mistrust.

CONCLUSIONS

A one-size approach to inform women about their BD may further disadvantage racial and ethnic minority women. BD notification and accompanying information should be tailored and translated to ensure readability and understandability by all women.

Contrast-Enhanced Mammography for Women with Palpable Breast Abnormalities

RATIONALE AND OBJECTIVES

To examine the role of contrast-enhanced mammography (CEM) in the work-up of palpable breast abnormalities.

MATERIALS AND METHODS

In this single-center combination prospective-retrospective study, women with palpable breast abnormalities underwent CEM evaluation prospectively, comprising the acquisition of low energy (LE) images and recombined images (RI) which depict enhancement, followed by targeted ultrasound (US). Two independent readers retrospectively reviewed the imaging and assigned BI-RADS assessment based on LE alone, LE plus US, RI with LE plus US (CEM plus US), and RI alone. Pathology results or 1-year follow-up imaging served as the reference standard.

RESULTS

237 women with 262 palpable abnormalities were included (mean age, 51 years). Of the 262 palpable abnormalities, 116/262 (44%) had no imaging correlate and 242/262 (92%) were benign. RI alone had better specificity compared to LE plus US (Reader 1, 94% versus 89% (p = 0.009); Reader 2, 93% versus 88% (p = 0.03)), better positive predictive value (Reader 1, 52% versus 42% (p = 0.04); Reader 2, 53% versus 42% (p = 0.04)), and better accuracy (Reader 1, 93% versus 89% (p = 0.05); Reader 2, 93% versus 90% (p = 0.06)). CEM plus US was not significantly different in performance metrics versus LE plus US.

CONCLUSION

RI had better specificity compared to LE in combination with US. There was no difference in performance between CEM plus US and LE plus US, likely reflecting the weight US carries in radiologist decision-making. However, the results indicate that the absence of enhancement on RI in the setting of palpable lesions may help avoid benign biopsies.

Comparative Effectiveness of Mammography, Ultrasound, and MRI in the Detection of Breast Carcinoma in Dense Breast Tissue: A Systematic Review

This systematic review aimed to critically assess the effectiveness of mammography, ultrasound, and magnetic resonance imaging (MRI) in the detection of breast carcinoma within dense breast tissue. An exhaustive search of contemporary literature was undertaken, focusing on the diagnostic accuracy, false positive and negative rates, and clinical implications of the aforementioned imaging modalities. Each modality was assessed in isolation and side by side against the others to draw comparative inferences. While mammography remains a foundational imaging modality, its effectiveness waned within the context of dense breast tissue. Ultrasound demonstrated a strong differentiation prowess, especially among specific demographic cohorts. MRI, despite its exceptional precision and differentiation capabilities, exhibited a tendency for slightly elevated false positive rates. No single modality emerged as singularly superior for all cases. Instead, an integrated approach, combining the strengths of each modality based on individual patient profiles and clinical scenarios, is recommended. This tailored approach ensures optimized detection rates and minimizes diagnostic ambiguities, underscoring the significance of individualized patient care in the field of diagnostic radiology.

Transfer learning for the generalization of artificial intelligence in breast cancer detection: a case-control study

BACKGROUND

Some researchers have questioned whether artificial intelligence (AI) systems maintain their performance when used for women from populations not considered during the development of the system.

PURPOSE

To evaluate the impact of transfer learning as a way of improving the generalization of AI systems in the detection of breast cancer.

MATERIAL AND METHODS

This retrospective case-control Finnish study involved 191 women diagnosed with breast cancer and 191 matched healthy controls. We selected a state-of-the-art AI system for breast cancer detection trained using a large US dataset. The selected baseline system was evaluated in two experimental settings. First, we examined our private Finnish sample as an independent test set that had not been considered in the development of the system (unseen population). Second, the baseline system was retrained to attempt to improve its performance in the unseen population by means of transfer learning. To analyze performance, we used areas under the receiver operating characteristic curve (AUCs) with DeLong's test.

RESULTS

Two versions of the baseline system were considered: ImageOnly and Heatmaps. The ImageOnly and Heatmaps versions yielded mean AUC values of 0.82±0.008 and 0.88±0.003 in the US dataset and 0.56 (95% CI=0.50-0.62) and 0.72 (95% CI=0.67-0.77) when evaluated in the unseen population, respectively. The retrained systems achieved AUC values of 0.61 (95% CI=0.55-0.66) and 0.69 (95% CI=0.64-0.75), respectively. There was no statistical difference between the baseline system and the retrained system.

CONCLUSION

Transfer learning with a small study sample did not yield a significant improvement in the generalization of the system.

Study on the differential diagnosis of benign and malignant breast lesions using a deep learning model based on multimodal images

OBJECTIVE

To establish a multimodal model for distinguishing benign and malignant breast lesions.

MATERIALS AND METHODS

Clinical data, mammography, and MRI images (including T2WI, diffusion-weighted images (DWI), apparent diffusion coefficient (ADC), and DCE-MRI images) of 132 benign and breast cancer patients were analyzed retrospectively. The region of interest (ROI) in each image was marked and segmented using MATLAB software. The mammography, T2WI, DWI, ADC, and DCE-MRI models based on the ResNet34 network were trained. Using an integrated learning method, the five models were used as a basic model, and voting methods were used to construct a multimodal model. The dataset was divided into a training set and a prediction set. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the model were calculated. The diagnostic efficacy of each model was analyzed using a receiver operating characteristic curve (ROC) and an area under the curve (AUC). The diagnostic value was determined by the DeLong test with statistically significant differences set at P < 0.05.

RESULTS

We evaluated the ability of the model to classify benign and malignant tumors using the test set. The AUC values of the multimodal model, mammography model, T2WI model, DWI model, ADC model and DCE-MRI model were 0.943, 0.645, 0.595, 0.905, 0.900, and 0.865, respectively. The diagnostic ability of the multimodal model was significantly higher compared with that of the mammography and T2WI models. However, compared with the DWI, ADC, and DCE-MRI models, there was no significant difference in the diagnostic ability of these models.

CONCLUSION

Our deep learning model based on multimodal image training has practical value for the diagnosis of benign and malignant breast lesions.

Breast Radiologists' Perceptions on the Detection and Management of Invasive Lobular Carcinoma: Most Agree Imaging Beyond Mammography Is Warranted

OBJECTIVE

To determine breast radiologists' confidence in detecting invasive lobular carcinoma (ILC) on mammography and the perceived need for additional imaging in screening and preoperative settings.

METHODS

A 16-item anonymized survey was developed, and IRB exemption obtained, by the Society of Breast Imaging (SBI) Patient Care and Delivery Committee and the Lobular Breast Cancer Alliance. The survey was emailed to 2946 radiologist SBI members on February 15, 2023. The survey recorded demographics, perceived modality-specific sensitivity for ILC to the nearest decile, and opinions on diagnosing ILC in screening and staging imaging. Five-point Likert scales were used (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree).

RESULTS

Response rate was 12.4% (366/2946). Perceived median (interquartile range) modality-specific sensitivities for ILC were MRI 90% (80-90), contrast-enhanced mammography 80% (70-90), molecular breast imaging 80% (60-90), digital breast tomosynthesis 70% (60-80), US 60% (50-80), and 2D mammography 50% (30-60). Only 25% (85/340) respondents were confident in detecting ILC on screening mammography in dense breasts, while 67% (229/343) were confident if breasts were nondense. Most agreed that supplemental screening is needed to detect ILC in women with dense breasts (272/344, 79%) or a personal history of ILC (248/341, 73%), with 34% (118/334) indicating that supplemental screening would also benefit women with nondense breasts. Most agreed that additional imaging is needed to evaluate extent of disease in women with newly diagnosed ILC, regardless of breast density (dense 320/329, 97%; nondense 263/329, 80%).

CONCLUSION

Most breast radiologists felt that additional imaging beyond mammography is needed to more confidently screen for and stage ILC.

Generalisable deep learning method for mammographic density prediction across imaging techniques and self-reported race

BACKGROUND

Breast density is an important risk factor for breast cancer complemented by a higher risk of cancers being missed during screening of dense breasts due to reduced sensitivity of mammography. Automated, deep learning-based prediction of breast density could provide subject-specific risk assessment and flag difficult cases during screening. However, there is a lack of evidence for generalisability across imaging techniques and, importantly, across race.

METHODS

This study used a large, racially diverse dataset with 69,697 mammographic studies comprising 451,642 individual images from 23,057 female participants. A deep learning model was developed for four-class BI-RADS density prediction. A comprehensive performance evaluation assessed the generalisability across two imaging techniques, full-field digital mammography (FFDM) and two-dimensional synthetic (2DS) mammography. A detailed subgroup performance and bias analysis assessed the generalisability across participants' race.

RESULTS

Here we show that a model trained on FFDM-only achieves a 4-class BI-RADS classification accuracy of 80.5% (79.7-81.4) on FFDM and 79.4% (78.5-80.2) on unseen 2DS data. When trained on both FFDM and 2DS images, the performance increases to 82.3% (81.4-83.0) and 82.3% (81.3-83.1). Racial subgroup analysis shows unbiased performance across Black, White, and Asian participants, despite a separate analysis confirming that race can be predicted from the images with a high accuracy of 86.7% (86.0-87.4).

CONCLUSIONS

Deep learning-based breast density prediction generalises across imaging techniques and race. No substantial disparities are found for any subgroup, including races that were never seen during model development, suggesting that density predictions are unbiased.

Head-to-head comparison of cone-beam breast computed tomography and mammography in the diagnosis of primary breast cancer: A systematic review and meta-analysis

INTRODUCTION

To compare the diagnostic performance of cone-beam breast computed tomography (CBBCT) and mammography (MG) in primary breast cancer.

METHODS

PubMed, Embase, Web of Science, China National Knowledge Infrastructure, WanFang DATA, and China Science and Technology Journal databases were searched comprehensively from inception to March 2023. Sensitivity and specificity were calculated using bivariate random-effects models, and a summary receiver operating characteristic (SROC) curve was constructed. Bivariate I2 statistics and meta-regression analyses were also performed. The differences in diagnostic performance between CBBCT and MG were analysed using Z-test statistics. Clinical utility was explored using Fagan's nomogram, and quality assessment was conducted utilising the Quality Assessment of Diagnostic Accuracy Studies-2 checklist.

RESULTS

The summary sensitivity and specificity for CBBCT in diagnosing primary breast cancer were 0.92 (95 % CI: 0.87-0.94) and 0.79 (95 % CI: 0.71-0.85), respectively, and the area under the curve (AUC) of the SROC was 0.93 (95 % CI: 0.90-0.95). For MG, the summary sensitivity and specificity were 0.77 (95 % CI: 0.69-0.83) and 0.75 (95 % CI: 0.66-0.82), respectively, with an AUC of 0.83 (95 % CI: 0.80-0.86). The Z-test revealed that the summary sensitivity of CBBCT was significantly higher than that of MG (P < 0.001). Additionally, the summary AUC of CBBCT was significantly higher than that of MG (P < 0.001).

CONCLUSION

The diagnostic performance of CBBCT for primary breast cancer was better than that of MG. However, the results of both the CBBCT and MG are based on studies with small sample sizes. Further studies with larger sample sizes and more comprehensive designs are required to address this issue.

Evaluating Recalibrating AI Models for Breast Cancer Diagnosis in a New Context: Insights from Transfer Learning, Image Enhancement and High-Quality Training Data Integration

This paper investigates the adaptability of four state-of-the-art artificial intelligence (AI) models to the Australian mammographic context through transfer learning, explores the impact of image enhancement on model performance and analyses the relationship between AI outputs and histopathological features for clinical relevance and accuracy assessment. A total of 1712 screening mammograms (n = 856 cancer cases and n = 856 matched normal cases) were used in this study. The 856 cases with cancer lesions were annotated by two expert radiologists and the level of concordance between their annotations was used to establish two sets: a 'high-concordances subset' with 99% agreement of cancer location and an 'entire dataset' with all cases included. The area under the receiver operating characteristic curve (AUC) was used to evaluate the performance of Globally aware Multiple Instance Classifier (GMIC), Global-Local Activation Maps (GLAM), I&H and End2End AI models, both in the pretrained and transfer learning modes, with and without applying the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm. The four AI models with and without transfer learning in the high-concordance subset outperformed those in the entire dataset. Applying the CLAHE algorithm to mammograms improved the performance of the AI models. In the high-concordance subset with the transfer learning and CLAHE algorithm applied, the AUC of the GMIC model was highest (0.912), followed by the GLAM model (0.909), I&H (0.893) and End2End (0.875). There were significant differences (p < 0.05) in the performances of the four AI models between the high-concordance subset and the entire dataset. The AI models demonstrated significant differences in malignancy probability concerning different tumour size categories in mammograms. The performance of AI models was affected by several factors such as concordance classification, image enhancement and transfer learning. Mammograms with a strong concordance with radiologists' annotations, applying image enhancement and transfer learning could enhance the accuracy of AI models.

Rethinking screening mammography in Japan: next-generation breast cancer screening through breast awareness and supplemental ultrasonography

Breast cancer mortality has not been reduced in Japan despite more than 20 years of population-based screening mammography. Screening mammography might not be suitable for Japanese women who often have dense breasts, thus decreasing mammography sensitivity because of masking. The J-START study showed that breast ultrasonography increases the sensitivity and the detection rate for early invasive cancers and lowers the rate of interval cancers for Japanese women in their 40 s. Breast awareness and breast cancer survival are directly correlated; however, breast awareness is not widely known in Japan. Next-generation breast cancer screening in Japan should consist of breast awareness campaigns for improving breast cancer literacy and supplemental breast ultrasonography to address the problem of false-negative mammograms attributable to dense breasts.

Pathological macroscopic evaluation of breast density versus mammographic breast density in breast cancer conserving surgery

Correlation between imaging and anatomopathological breast density has been superficially explored and is heterogeneous in current medical literature. It is possible that mammographic and pathological findings are divergent. The aim of this study is to evaluate the association between breast density classified by mammography and breast density of pathological macroscopic examination in specimens of breast cancer conservative surgeries. Post-hoc, exploratory analysis of a prospective randomized clinical trial of patients with breast cancer candidates for breast conservative surgery. Breast mammographic density (MD) was analyzed according to ACR BI-RADS® criteria, and pathologic macroscopic evaluation of breast density (PMBD) was estimated by visually calculating the ratio between stromal and fatty tissue. From 412 patients, MD was A in 291 (70,6%), B in 80 (19,4%) B, C in 35 (8,5%), and D in 6 (1,5%). Ninety-nine percent (201/203) of patients classified as A+B in MD were correspondently classified in PMBD. Conversely, only 18.7% (39/209) of patients with MD C+D were classified correspondently in PMBD (p < 0.001). Binary logistic regression showed age (OR 1.06, 1.01-1.12 95% CI, p 0.013) and nulliparity (OR 0.39, 0.17-0.96 95% CI, p 0.039) as predictors of A+B PMBD.

Conclusion

Mammographic and pathologic macroscopic breast density showed no association in our study for breast C or D in breast image. The fatty breast was associated with older patients and the nulliparity decreases the chance of fatty breasts nearby 60%.

The Level of Awareness Among Healthcare Practitioners Regarding the Relationship Between Breast Density and Breast Cancer

Background Breast cancer is the most prevalent cancer in women, accounting for around 23% of all cancer-related deaths across 140 nations. The awareness about breast density (BD) has a significant impact on early diagnosis of breast cancer. Aim and objective This study aims to assess the awareness of healthcare providers about BD in King Abdullah Medical City. Methods This is an analytical cross-sectional questionnaire-based study among the healthcare practitioners of KAMC in Makkah, Saudi Arabia. Questions measured knowledge about BD and a pass mark indicated participant awareness. The collected data were analyzed using SPSS, and a chi-square test used for bivariate analysis. Results Out of 124 participants, 41% were well aware. Physicians (37% of the sample) were significantly more aware than allied healthcare practitioners and nurses (awareness: 59.6%, 33.3%, 30.4% respectively, (p = 0.03)). Regarding specialty, radiologists and surgeons had the top level of awareness (62% and 64%, respectively) as compared to oncologists (47.1%) and other specialties (29.7%), (p= 0.016). Those above 40 years of age were more aware than those below 40 years (awareness: 62.1% and 34%, respectively, (p=0.007)). Non-significant factors included: gender, years of experience, screened versus non-screened, and receiving information before about BD (p > 0.05). Conclusion The results of this population-based study indicate the existence of moderate deficits in the general knowledge about BD and its relation to breast cancer. This might lead to a late diagnosis. The results showed no dramatic differences in the awareness among healthcare providers.

Breast Cancer Detection with an Ensemble of Deep Learning Networks Using a Consensus-Adaptive Weighting Method

Breast cancer's high mortality rate is often linked to late diagnosis, with mammograms as key but sometimes limited tools in early detection. To enhance diagnostic accuracy and speed, this study introduces a novel computer-aided detection (CAD) ensemble system. This system incorporates advanced deep learning networks-EfficientNet, Xception, MobileNetV2, InceptionV3, and Resnet50-integrated via our innovative consensus-adaptive weighting (CAW) method. This method permits the dynamic adjustment of multiple deep networks, bolstering the system's detection capabilities. Our approach also addresses a major challenge in pixel-level data annotation of faster R-CNNs, highlighted in a prominent previous study. Evaluations on various datasets, including the cropped DDSM (Digital Database for Screening Mammography), DDSM, and INbreast, demonstrated the system's superior performance. In particular, our CAD system showed marked improvement on the cropped DDSM dataset, enhancing detection rates by approximately 1.59% and achieving an accuracy of 95.48%. This innovative system represents a significant advancement in early breast cancer detection, offering the potential for more precise and timely diagnosis, ultimately fostering improved patient outcomes.

Psychosocial outcomes and health service use after notifying women participating in population breast screening when they have dense breasts: a BreastScreen Queensland randomised controlled trial

BACKGROUND

Robust evidence regarding the benefits and harms of notifying Australian women when routine breast screening identifies that they have dense breasts is needed for informing future mammography population screening practice and policy.

OBJECTIVES

To assess the psychosocial and health services use effects of notifying women participating in population-based breast cancer screening that they have dense breasts; to examine whether the mode of communicating this information about its implications (print, online formats) influences these effects.

METHODS AND ANALYSIS

The study population comprises women aged 40 years or older who attend BreastScreen Queensland Sunshine Coast services for mammographic screening and are found to have dense breasts (BI-RADS density C or D). The randomised controlled trial includes three arms (952 women each): standard BreastScreen care (no notification of breast density; control arm); notification of dense breasts in screening results letter and print health literacy-sensitive information (intervention arm 1) or a link or QR code to online video-based health literacy-sensitive information (intervention arm 2). Baseline demographic data will be obtained from BreastScreen Queensland. Outcomes data will be collected in questionnaires at baseline and eight weeks, twelve months, and 27 months after breast screening. Primary outcomes will be psychological outcomes and health service use; secondary outcomes will be supplemental screening outcomes, cancer worry, perceived breast cancer risk, knowledge about breast density, future mammographic screening intentions, and acceptability of notification about dense breasts.

ETHICS APPROVAL

Gold Coast Hospital and Health Service Ethics Committee (HREC/2023/QGC/89770); Sunshine Coast Hospital and Health Service Research Governance and Development (SSA/2023/QSC/89770).

DISSEMINATION OF FINDINGS

Findings will be reported in peer-reviewed journals and at national and international conferences. They will also be reported to BreastScreen Queensland, BreastScreen Australia, Cancer Australia, and other bodies involved in cancer care and screening, including patient and support organisations.

TRIAL REGISTRATION

Australian New Zealand Clinical Trials Registry ACTRN12623000001695p (prospective: 9 January 2023).

Breast density and the likelihood of malignant MRI-detected lesions in women diagnosed with breast cancer

OBJECTIVES

To assess whether mammographic breast density in women diagnosed with breast cancer correlates with the total number of incidental magnetic resonance imaging (MRI)-detected lesions and the likelihood of the lesions being malignant.

METHODS

Patients diagnosed with breast cancer meeting the EUSOBI and EUSOMA criteria for preoperative breast MRI routinely undergo mammography and ultrasound before MRI at our institution. Incidental suspicious breast lesions detected in MRI are biopsied. We included patients diagnosed with invasive breast cancers between 2014 and 2019 who underwent preoperative breast MRI. One reader retrospectively determined breast density categories according to the 5th edition of the BI-RADS lexicon.

RESULTS

Of 946 patients with 973 malignant primary breast tumors, 166 (17.5%) had a total of 175 (18.0%) incidental MRI-detected lesions (82 (46.9%) malignant and 93 (53.1%) benign). High breast density according to BI-RADS was associated with higher incidence of all incidental enhancing lesions in preoperative breast MRIs: 2.66 (95% confidence interval: 1.03-6.86) higher for BI-RADS density category B, 2.68 (1.04-6.92) for category C, and 3.67 (1.36-9.93) for category D compared to category A (p < 0.05). However, high breast density did not predict higher incidence of malignant incidental lesions (p = 0.741). Incidental MRI-detected lesions in the contralateral breast were more likely benign (p < 0.001): 18 (27.3%)/48 (72.7%) vs. 64 (58.7%)/45 (41.3%) malignant/benign incidental lesions in contralateral vs. ipsilateral breasts.

CONCLUSION

Women diagnosed with breast cancer who have dense breasts have more incidental MRI-detected lesions, but higher breast density does not translate to increased likelihood of malignant incidental lesions.

CLINICAL RELEVANCE STATEMENT

Dense breasts should not be considered as an indication for preoperative breast MRI in women diagnosed with breast cancer.

KEY POINTS

• The role of preoperative MRI of patients with dense breasts diagnosed with breast cancer is under debate. • Women with denser breasts have a higher incidence of all MRI-detected incidental breast lesions, but the incidence of malignant MRI-detected incidental lesions is not higher than in women with fatty breasts. • High breast density alone should not indicate preoperative breast MRI.

Impact of BMI on Prevalence of Dense Breasts by Race and Ethnicity

BACKGROUND

Density notification laws require notifying women of dense breasts with dense breast prevalence varying by race/ethnicity. We evaluated whether differences in body mass index (BMI) account for differences in dense breasts prevalence by race/ethnicity.

METHODS

Prevalence of dense breasts (heterogeneously or extremely dense) according to Breast Imaging Reporting and Data System and obesity (BMI > 30 kg/m2) were estimated from 2,667,207 mammography examinations among 866,033 women in the Breast Cancer Surveillance Consortium (BCSC) from January 2005 through April 2021. Prevalence ratios (PR) for dense breasts relative to overall prevalence by race/ethnicity were estimated by standardizing race/ethnicity prevalence in the BCSC to the 2020 U.S. population, and adjusting for age, menopausal status, and BMI using logistic regression.

RESULTS

Dense breasts were most prevalent among Asian women (66.0%) followed by non-Hispanic/Latina (NH) White (45.5%), Hispanic/Latina (45.3%), and NH Black (37.0%) women. Obesity was most prevalent in Black women (58.4%) followed by Hispanic/Latina (39.3%), NH White (30.6%), and Asian (8.5%) women. The adjusted prevalence of dense breasts was 19% higher [PR = 1.19; 95% confidence interval (CI), 1.19-1.20] in Asian women, 8% higher (PR = 1.08; 95% CI, 1.07-1.08) in Black women, the same in Hispanic/Latina women (PR = 1.00; 95% CI, 0.99-1.01), and 4% lower (PR = 0.96; 95% CI, 0.96-0.97) in NH White women relative to the overall prevalence.

CONCLUSIONS

Clinically important differences in breast density prevalence are present across racial/ethnic groups after accounting for age, menopausal status, and BMI.

IMPACT

If breast density is the sole criterion used to notify women of dense breasts and discuss supplemental screening it may result in implementing inequitable screening strategies across racial/ethnic groups. See related In the Spotlight, p. 1479.

Ultra Sensitive Breast Cancer Cell Lines Detection Using Dual Nanocavities Engraved Junctionless FET

This article reports breast cancer cell lines (Hs578T, MDA-MB-231, MCF-7, and T47D) and healthy breast cells (MCF-10A) detection based on the modulation of its electrical properties by deploying dual nanocavity engraved junctionless FET. The device has a dual gate to enhance gate control and has two nanocavities etched under both gates for breast cancer cell lines immobilization. As the cancer cells are immobilized in the engraved nanocavities, which were earlier filled with air, the dielectric constant of the nanocavities changes. This results in the modulation of the device's electrical parameters. This electrical parameters modulation is then calibrated to detect the breast cancer cell lines. The reported device demonstrates a higher sensitivity toward the detection of breast cancer cells. The JLFET device optimization is done for improving the performance by optimizing the nanocavity thickness and the SiO2 oxide length. The variation in the dielectric property of cell lines plays a key role in the detection technique of the reported biosensor. The sensitivity of the JLFET biosensor is analyzed in terms of ∆VTH, ∆ION, ∆gm , and ∆SS . The reported biosensor shows the maximum sensitivity for T47D ( κ = 32 ) breast cancer cell line with ∆VTH = 0.800 V, ∆ION = 0.165 mA/μm, ∆gm = 0.296 mA/V-μm , and ∆SS = 5.41 mV/decade. Moreover, the effect of variation in the occupancy of the cavity by the immobilized cell lines has also been studied and analyzed. With increased cavity occupancy the variation in the device performance parameter enhances Further, the sensitivity of the proposed biosensor is compared with the existing biosensors and it is reported to be highly sensitive as compared to the existing biosensors. Hence, the device can be utilized for array based screening of cell lines of breast cancer and diagnosis with the benefit of easier fabrication and cost effectiveness of the device.

Association of polycystic ovary syndrome with mammographic density in Turkish women: a population-based case-control study

OBJECTIVE

The objective of this study was to investigate the breast densities and Breast Imaging-Reporting and Data System scores of patients with polycystic ovary syndrome and normoovulatory women and to determine whether these patients constitute a high-risk population for breast cancer.

METHODS

This retrospective case-control study was conducted at our institution between January 2022 and December 2022, involving patients diagnosed with polycystic ovary syndrome. Menstrual periods, hyperandrogenemic findings, and ultrasound reports of the patients were retrieved from our hospital's database. Patients who met at least two of the Rotterdam criteria were included in the polycystic ovary syndrome group. A total of 70 premenopausal patients over the age of 40 years, diagnosed with polycystic ovary syndrome, and 70 normoovulatory women, matched for age and body mass index, were included in the study. The two groups were compared regarding age at menarche, menstrual pattern, gravida, parity, levels of follicle-stimulating hormone, luteinizing hormone, and estradiol, endometrial thickness, breast density category, and Breast Imaging-Reporting and Data System classifications.

RESULTS

Patients in the polycystic ovary syndrome group had a higher age at menarche (12.7 vs. 12.3, p=0.006). There was no difference between the gonadotropin levels in both groups. However, the estradiol level was higher in the polycystic ovary syndrome group (p<0.001). There was no statistically significant difference between the two groups in terms of breast density and Breast Imaging-Reporting and Data System scores (p=0.319 and p=0.650, respectively).

CONCLUSION

Although we can conclude that the risk of breast malignancy is not increased in patients with polycystic ovary syndrome, the impact of the complex hormonal status of polycystic ovary syndrome on breast cancer remains unclear in the literature.

A review of the machine learning datasets in mammography, their adherence to the FAIR principles and the outlook for the future

The increasing rates of breast cancer, particularly in emerging economies, have led to interest in scalable deep learning-based solutions that improve the accuracy and cost-effectiveness of mammographic screening. However, such tools require large volumes of high-quality training data, which can be challenging to obtain. This paper combines the experience of an AI startup with an analysis of the FAIR principles of the eight available datasets. It demonstrates that the datasets vary considerably, particularly in their interoperability, as each dataset is skewed towards a particular clinical use-case. Additionally, the mix of digital captures and scanned film compounds the problem of variability, along with differences in licensing terms, ease of access, labelling reliability, and file formats. Improving interoperability through adherence to standards such as the BIRADS criteria for labelling and annotation, and a consistent file format, could markedly improve access and use of larger amounts of standardized data. This, in turn, could be increased further by GAN-based synthetic data generation, paving the way towards better health outcomes for breast cancer.

Can the Clinical Utility of iBreastExam, a Novel Device, Aid in Optimizing Breast Cancer Diagnosis? A Systematic Review

PURPOSE

A portable, cost-effective, easy-to-use, hand-held Intelligent Breast Exam (iBE), which is a wireless, radiation-free device, may be a valuable screening tool in resource-limited settings. While multiple studies evaluating the use of iBE have been conducted worldwide, there are no cumulative studies evaluating the iBE's performance. Therefore this review aims to determine the clinical utility and applicability of iBE compared with clinical breast examinations, ultrasound, and mammography and discuss its strengths and weaknesses when performing breast-cancer screening.

METHODS

A systematic review was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Four electronic databases were searched: PubMed, Cochrane Library, Web of Science, and Google Scholar.

RESULTS

The review included 11 studies with a total sample size of 16,052 breasts. The mean age ranged from 42 to 58 years. The sensitivity and specificity of the iBE ranged from 34.3% to 86% and 59% to 94%, respectively. For malignant lesions, iBE demonstrated a moderate to higher diagnostic capacity ranging from 57% to 93% and could identify tumor sizes spanning from 0.5 cm to 9 cm.

CONCLUSION

Our findings underscore the potential clinical utility and applicability of iBE as a prescreening and triaging tool, which may aid in reducing the burden of patients undergoing diagnostic imaging in lower- and middle-income countries. Furthermore, iBE has shown to diagnose cancers as small as 0.5 cm, which can be a boon in early detection and reduce mortality rates. However, the encouraging results of this systematic review should be interpreted with caution because of the device's low sensitivity and high false-positive rates.

Interval Breast Cancers Versus Screen Detected Breast Cancers: A Retrospective Cohort Study

RATIONALE AND OBJECTIVE

Mammographic screening detects most breast cancers but there are still women diagnosed with breast cancer between annual mammograms. We aim to identify features that differentiate screen detected breast cancers from interval breast cancer.

MATERIALS AND METHODS

All screening mammograms (n = 211,517) performed 7/1/2013-6/30/2020 at our institution were reviewed. Patients with breast cancer diagnosed within one year of screening were included and divided into two distinct groups: screen detected cancer group and interval cancer group. Characteristics in these groups were compared using the chi square test, fisher test, and student's T test.

RESULTS

A total of 1,232 patients were included (mean age 64 +/- 11). Sensitivity of screening mammography was 92% (1,136 screen detected cancers, 96 interval cancers). Patient age, race, and personal history of breast cancer were similar between the groups (p > 0.05). Patients with interval cancers more often had dense breast tissue (75/96 = 78% versus 694/1136 = 61%, p < 0.001). Compared to screen detected cancers, interval cancers were more often primary tumor stage two or higher (41/96 = 43% versus 139/1136 = 12%, p < 0.001) and regional lymph node stage one or higher (21/96 = 22% versus 132/1136 = 12%, p = 0.003). Interval cancers were more often triple negative (16/77 = 21% versus [48/813 = 6%], p < 0.001) with high Ki67 proliferation indices (28/45 = 62% versus 188/492 = 38%, p = 0.002).

CONCLUSION

Mammographic screening had high sensitivity for breast cancer detection (92%). Interval cancers were associated with dense breast tissue and had higher stage with less favorable molecular features compared to screen detected cancers.

An integrated network based on 2D/3D feature correlations for benign-malignant tumor classification and uncertainty estimation in digital breast tomosynthesis

Objective.Classification of benign and malignant tumors is important for the early diagnosis of breast cancer. Over the last decade, digital breast tomosynthesis (DBT) has gradually become an effective imaging modality for breast cancer diagnosis due to its ability to generate three-dimensional (3D) visualizations. However, computer-aided diagnosis (CAD) systems based on 3D images require high computational costs and time. Furthermore, there is considerable redundant information in 3D images. Most CAD systems are designed based on 2D images, which may lose the spatial depth information of tumors. In this study, we propose a 2D/3D integrated network for the diagnosis of benign and malignant breast tumors.Approach.We introduce a correlation strategy to describe feature correlations between slices in 3D volumes, corresponding to the tissue relationship and spatial depth features of tumors. The correlation strategy can be used to extract spatial features with little computational cost. In the prediction stage, 3D spatial correlation features and 2D features are both used for classification.Main results.Experimental results demonstrate that our proposed framework achieves higher accuracy and reliability than pure 2D or 3D models. Our framework has a high area under the curve of 0.88 and accuracy of 0.82. The parameter size of the feature extractor in our framework is only 35% of that of the 3D models. In reliability evaluations, our proposed model is more reliable than pure 2D or 3D models because of its effective and nonredundant features.Significance.This study successfully combines 3D spatial correlation features and 2D features for the diagnosis of benign and malignant breast tumors in DBT. In addition to high accuracy and low computational cost, our model is more reliable and can output uncertainty value. From this point of view, the proposed method has the potential to be applied in clinic.

Two-stage classification strategy for breast cancer diagnosis using ultrasound-guided diffuse optical tomography and deep learning

Significance

Ultrasound (US)-guided diffuse optical tomography (DOT) has demonstrated great potential for breast cancer diagnosis in which real-time or near real-time diagnosis with high accuracy is desired.

Aim

We aim to use US-guided DOT to achieve an automated, fast, and accurate classification of breast lesions.

Approach

We propose a two-stage classification strategy with deep learning. In the first stage, US images and histograms created from DOT perturbation measurements are combined to predict benign lesions. Then the non-benign suspicious lesions are passed through to the second stage, which combine US image features, DOT histogram features, and 3D DOT reconstructed images for final diagnosis.

Results

The first stage alone identified 73.0% of benign cases without image reconstruction. In distinguishing between benign and malignant breast lesions in patient data, the two-stage classification approach achieved an area under the receiver operating characteristic curve of 0.946, outperforming the diagnoses of all single-modality models and of a single-stage classification model that combines all US images, DOT histogram, and imaging features.

Conclusions

The proposed two-stage classification strategy achieves better classification accuracy than single-modality-only models and a single-stage classification model that combines all features. It can potentially distinguish breast cancers from benign lesions in near real-time.

Deep Learning Analysis of Mammography for Breast Cancer Risk Prediction in Asian Women

The purpose of this study was to develop a mammography-based deep learning (DL) model for predicting the risk of breast cancer in Asian women. This retrospective study included 287 examinations in 153 women in the cancer group and 736 examinations in 447 women in the negative group, obtained from the databases of two tertiary hospitals between November 2012 and March 2022. All examinations were labeled as either dense breast or nondense breast, and then randomly assigned to either training, validation, or test sets. DL models, referred to as image-level and examination-level models, were developed. Both models were trained to predict whether or not the breast would develop breast cancer with two datasets: the whole dataset and the dense-only dataset. The performance of DL models was evaluated using the accuracy, precision, sensitivity, specificity, F1 score, and area under the receiver operating characteristic curve (AUC). On a test set, performance metrics for the four scenarios were obtained: image-level model with whole dataset, image-level model with dense-only dataset, examination-level model with whole dataset, and examination-level model with dense-only dataset with AUCs of 0.71, 0.75, 0.66, and 0.67, respectively. Our DL models using mammograms have the potential to predict breast cancer risk in Asian women.

Imaging of breast cancer-beyond the basics

Imaging of breast cancer is the backbone of breast cancer screening, diagnosis, preoperative/treatment assessment and follow-up. The main modalities are mammography, ultrasound and magnetic resonance imaging, each with its own advantages and disadvantages. New emerging technologies have also enabled each modality to improve on their weaknesses. Imaging-guided biopsies have allowed for accurate diagnosis of breast cancer, with low complication rates. The purpose of this article is to review the common modalities for breast cancer imaging in current practice with emphasis on the strengths and potential weaknesses, discuss the selection of the best imaging modality for the specific clinical question or patient population, and explore new technologies / future directions of breast cancer imaging.

ACR Appropriateness Criteria® Palpable Breast Masses: 2022 Update

Palpable masses in women are the most common symptom associated with breast cancer. This document reviews and evaluates the current evidence for imaging recommendations of palpable masses in women less than 30 to over 40 years of age. There is also a review of several different scenarios and recommendations after initial imaging. Ultrasound is usually the appropriate initial imaging for women under 30 years of age. If ultrasound findings are suspicious or highly suggestive of malignancy (BIRADS 4 or 5), it is usually appropriate to continue with diagnostic tomosynthesis or mammography with image-guided biopsy. No further imaging is recommended if the ultrasound is benign or negative. The patient under 30 years of age with a probably benign ultrasound may undergo further imaging; however, the clinical scenario plays a role in the decision to biopsy. For women between 30 to 39 years of age, ultrasound, diagnostic mammography, tomosynthesis, and ultrasound are usually appropriate. Diagnostic mammography and tomosynthesis are the appropriate initial imaging for women 40 years of age or older, as ultrasound may be appropriate if the patient had a negative mammogram within 6 months of presentation or immediately after mammography findings are suspicious or highly suggestive of malignancy. If the diagnostic mammogram, tomosynthesis, and ultrasound findings are probably benign, no further imaging is necessary unless the clinical scenario indicates a biopsy. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision process support the systematic analysis of the medical literature from peer reviewed journals. Established methodology principles such as Grading of Recommendations Assessment, Development, and Evaluation or GRADE are adapted to evaluate the evidence. The RAND/UCLA Appropriateness Method User Manual provides the methodology to determine the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where peer reviewed literature is lacking or equivocal, experts may be the primary evidentiary source available to formulate a recommendation.

Women 75 Years Old or Older: To Screen or Not to Screen?

Breast cancer is the most common cancer in women, with the incidence rising substantially with age. Older women are a vulnerable population at increased risk of developing and dying from breast cancer. However, women aged 75 years and older were excluded from all randomized controlled screening trials, so the best available data regarding screening benefits and risks in this age group are from observational studies and modeling predictions. Benefits of screening in older women are the same as those in younger women: early detection of smaller lower-stage cancers, resulting in less invasive treatment and lower morbidity and mortality. Mammography performs significantly better in older women with higher sensitivity, specificity, cancer detection rate, and positive predictive values, accompanied by lower recall rates and false positives. The overdiagnosis rate is low, with benefits outweighing risks until age 90 years. Although there are conflicting national and international guidelines about whether to continue screening mammography in women beyond age 74 years, clinicians can use shared decision making to help women make decisions about screening and fully engage them in the screening process. For women aged 75 years and older in good health, continuing annual screening mammography will save the most lives. An informed discussion of the benefits and risks of screening mammography in older women needs to include each woman's individual values, overall health status, and comorbidities. This article will review the benefits, risks, and controversies surrounding screening mammography in women 75 years old and older and compare the current recommendations for screening this population from national and international professional organizations. ^©RSNA, 2023 Quiz questions for this article are available through the Online Learning Center.

Predicting work disability among people with chronic conditions: a prospective cohort study

Few risk prediction scores are available to identify people at increased risk of work disability, particularly for those with an existing morbidity. We examined the predictive performance of disability risk scores for employees with chronic disease. We used prospective data from 88,521 employed participants (mean age 43.1) in the Finnish Public Sector Study including people with chronic disorders: musculoskeletal disorder, depression, migraine, respiratory disease, hypertension, cancer, coronary heart disease, diabetes, comorbid depression and cardiometabolic disease. A total of 105 predictors were assessed at baseline. During a mean follow-up of 8.6 years, 6836 (7.7%) participants were granted a disability pension. C-statistics for the 8-item Finnish Institute of Occupational Health (FIOH) risk score, comprising age, self-rated health, number of sickness absences, socioeconomic position, number of chronic illnesses, sleep problems, BMI, and smoking at baseline, exceeded 0.72 for all disease groups and was 0.80 (95% CI 0.80-0.81) for participants with musculoskeletal disorders, 0.83 (0.82-0.84) for those with migraine, and 0.82 (0.81-0.83) for individuals with respiratory disease. Predictive performance was not significantly improved in models with re-estimated coefficients or a new set of predictors. These findings suggest that the 8-item FIOH work disability risk score may serve as a scalable screening tool in identifying individuals with increased risk for work disability.

Radiologists' performance in reading digital breast tomosynthesis with and without synthesized views for cancer detection

OBJECTIVE

The study aims to evaluate the diagnostic efficacy of radiologists and radiology trainees in digital breast tomosynthesis (DBT) alone vs DBT plus synthesized view (SV) for an understanding of the adequacy of DBT images to identify cancer lesions.

METHODS

Fifty-five observers (30 radiologists and 25 radiology trainees) participated in reading a set of 35 cases (15 cancer) with 28 readers reading DBT and 27 readers reading DBT plus SV. Two groups of readers had similar experience in interpreting mammograms. The performances of participants in each reading mode were compared with the ground truth and calculated in term of specificity, sensitivity, and ROC AUC. The cancer detection rate in various levels of breast density, lesion types and lesion sizes between 'DBT' and 'DBT + SV' were also analyzed. The difference in diagnostic accuracy of readers between two reading modes was assessed using Man-Whitney U test. p < 0.05 indicated a significant result.

RESULTS

There was no significant difference in specificity (0.67-vs-0.65; p = 0.69), sensitivity (0.77-vs-0.71; p = 0.09), ROC AUC (0.77-vs-0.73; p = 0.19) of radiologists reading DBT plus SV compared with radiologists reading DBT. Similar result was found in radiology trainees with no significant difference in specificity (0.70-vs-0.63; p = 0.29), sensitivity (0.44-vs-0.55; p = 0.19), ROC AUC (0.59-vs-0.62; p = 0.60) between two reading modes. Radiologists and trainees obtained similar results in two reading modes for cancer detection rate with different levels of breast density, cancer types and sizes of lesions (p > 0.05).

CONCLUSION

Findings show that the diagnostic performances of radiologists and radiology trainees in DBT alone and DBT plus SV were equivalent in identifying cancer and normal cases.

ADVANCES IN KNOWLEDGE

DBT alone had equivalent diagnostic accuracy as DBT plus SV which could imply the consideration of using DBT as a sole modality without SV.

Mammography in combination with breast ultrasonography versus mammography for breast cancer screening in women at average risk

BACKGROUND

Screening mammography can detect breast cancer at an early stage. Supporters of adding ultrasonography to the screening regimen consider it a safe and inexpensive approach to reduce false-negative rates during screening. However, those opposed to it argue that performing supplemental ultrasonography will also increase the rate of false-positive findings and can lead to unnecessary biopsies and treatments.

OBJECTIVES

To assess the comparative effectiveness and safety of mammography in combination with breast ultrasonography versus mammography alone for breast cancer screening for women at average risk of breast cancer.

SEARCH METHODS

We searched the Cochrane Breast Cancer Group's Specialised Register, CENTRAL, MEDLINE, Embase, the World Health Organization International Clinical Trials Registry Platform (WHO ICTRP), and ClinicalTrials.gov up until 3 May 2021.

SELECTION CRITERIA

For efficacy and harms, we considered randomised controlled trials (RCTs) and controlled non-randomised studies enrolling at least 500 women at average risk for breast cancer between the ages of 40 and 75. We also included studies where 80% of the population met our age and breast cancer risk inclusion criteria.

DATA COLLECTION AND ANALYSIS

Two review authors screened abstracts and full texts, assessed risk of bias, and applied the GRADE approach. We calculated the risk ratio (RR) with 95% confidence intervals (CI) based on available event rates. We conducted a random-effects meta-analysis.

MAIN RESULTS

We included eight studies: one RCT, two prospective cohort studies, and five retrospective cohort studies, enrolling 209,207 women with a follow-up duration from one to three years. The proportion of women with dense breasts ranged from 48% to 100%. Five studies used digital mammography; one study used breast tomosynthesis; and two studies used automated breast ultrasonography (ABUS) in addition to mammography screening. One study used digital mammography alone or in combination with breast tomosynthesis and ABUS or handheld ultrasonography. Six of the eight studies evaluated the rate of cancer cases detected after one screening round, whilst two studies screened women once, twice, or more. None of the studies assessed whether mammography screening in combination with ultrasonography led to lower mortality from breast cancer or all-cause mortality. High certainty evidence from one trial showed that screening with a combination of mammography and ultrasonography detects more breast cancer than mammography alone. The J-START (Japan Strategic Anti-cancer Randomised Trial), enrolling 72,717 asymptomatic women, had a low risk of bias and found that two additional breast cancers per 1000 women were detected over two years with one additional ultrasonography than with mammography alone (5 versus 3 per 1000; RR 1.54, 95% CI 1.22 to 1.94). Low certainty evidence showed that the percentage of invasive tumours was similar, with no statistically significant difference between the two groups (69.6% (128 of 184) versus 73.5% (86 of 117); RR 0.95, 95% CI 0.82 to 1.09). However, positive lymph node status was detected less frequently in women with invasive cancer who underwent mammography screening in combination with ultrasonography than in women who underwent mammography alone (18% (23 of 128) versus 34% (29 of 86); RR 0.53, 95% CI 0.33 to 0.86; moderate certainty evidence). Further, interval carcinomas occurred less frequently in the group screened by mammography and ultrasonography compared with mammography alone (5 versus 10 in 10,000 women; RR 0.50, 95% CI 0.29 to 0.89; 72,717 participants; high certainty evidence). False-negative results were less common when ultrasonography was used in addition to mammography than with mammography alone: 9% (18 of 202) versus 23% (35 of 152; RR 0.39, 95% CI 0.23 to 0.66; moderate certainty evidence). However, the number of false-positive results and necessary biopsies were higher in the group with additional ultrasonography screening. Amongst 1000 women who do not have cancer, 37 more received a false-positive result when they participated in screening with a combination of mammography and ultrasonography than with mammography alone (RR 1.43, 95% CI 1.37 to 1.50; high certainty evidence). Compared to mammography alone, for every 1000 women participating in screening with a combination of mammography and ultrasonography, 27 more women will have a biopsy (RR 2.49, 95% CI 2.28 to 2.72; high certainty evidence). Results from cohort studies with methodological limitations confirmed these findings. A secondary analysis of the J-START provided results from 19,213 women with dense and non-dense breasts. In women with dense breasts, the combination of mammography and ultrasonography detected three more cancer cases (0 fewer to 7 more) per 1000 women screened than mammography alone (RR 1.65, 95% CI 1.0 to 2.72; 11,390 participants; high certainty evidence). A meta-analysis of three cohort studies with data from 50,327 women with dense breasts supported this finding, showing that mammography and ultrasonography combined led to statistically significantly more diagnosed cancer cases compared to mammography alone (RR 1.78, 95% CI 1.23 to 2.56; 50,327 participants; moderate certainty evidence). For women with non-dense breasts, the secondary analysis of the J-START study demonstrated that more cancer cases were detected when adding ultrasound to mammography screening compared to mammography alone (RR 1.93, 95% CI 1.01 to 3.68; 7823 participants; moderate certainty evidence), whilst two cohort studies with data from 40,636 women found no statistically significant difference between the two screening methods (RR 1.13, 95% CI 0.85 to 1.49; low certainty evidence).

AUTHORS' CONCLUSIONS

Based on one study in women at average risk of breast cancer, ultrasonography in addition to mammography leads to more screening-detected breast cancer cases. For women with dense breasts, cohort studies more in line with real-life clinical practice confirmed this finding, whilst cohort studies for women with non-dense breasts showed no statistically significant difference between the two screening interventions. However, the number of false-positive results and biopsy rates were higher in women receiving additional ultrasonography for breast cancer screening. None of the included studies analysed whether the higher number of screen-detected cancers in the intervention group resulted in a lower mortality rate compared to mammography alone. Randomised controlled trials or prospective cohort studies with a longer observation period are needed to assess the effects of the two screening interventions on morbidity and mortality.

Breast Density Evaluation According to BI-RADS 5th Edition on Digital Breast Tomosynthesis: AI Automated Assessment Versus Human Visual Assessment

Background: The assessment of breast density is one of the main goals of radiologists because the masking effect of dense fibroglandular tissue may affect the mammographic identification of lesions. The BI-RADS 5th Edition has revised the mammographic breast density categories, focusing on a qualitative evaluation rather than a quantitative one. Our purpose is to compare the concordance of the automatic classification of breast density with the visual assessment according to the latest available classification. Methods: A sample of 1075 digital breast tomosynthesis images from women aged between 40 and 86 years (58 ± 7.1) was retrospectively analyzed by three independent readers according to the BI-RADS 5th Edition. Automated breast density assessment was performed on digital breast tomosynthesis images with the Quantra software version 2.2.3. Interobserver agreement was assessed with kappa statistics. The distributions of breast density categories were compared and correlated with age. Results: The agreement on breast density categories was substantial to almost perfect between radiologists (κ = 0.63–0.83), moderate to substantial between radiologists and the Quantra software (κ = 0.44–0.78), and the consensus of radiologists and the Quantra software (κ = 0.60–0.77). Comparing the assessment for dense and non-dense breasts, the agreement was almost perfect in the screening age range without a statistically significant difference between concordant and discordant cases when compared by age. Conclusions: The categorization proposed by the Quantra software has shown a good agreement with the radiological evaluations, even though it did not completely reflect the visual assessment. Thus, clinical decisions regarding supplemental screening should be based on the radiologist’s perceived masking effect rather than the data produced exclusively by the Quantra software.

Validity of computed mean compressed fibroglandular tissue thickness and breast composition for stratification of masking risk in Japanese women

BACKGROUND

The volumetric measurement system for mammographic breast density is a high-precision objective method for evaluating the percentage of fibroglandular tissue volume (FG%). Nonetheless, FG% does not precisely correlate with subjective visual estimation (SVE) and shows poor evaluation performance regarding masking risk in patients with comparatively thin compressed breast thickness (CBT), commonly found in Japanese women. We considered that the mean compressed fibroglandular tissue thickness (mCGT), which incorporates the CBT element into the evaluation of breast density, may better predict masking risk.

METHODS

Volumetric measurements and SVEs were performed on mammograms of 108 breast cancer patients from our center. mCGT was calculated as the product of CBT and FG%. SVE was classified using the Breast Imaging-Reporting and Data System classification, 5th edition. Subsequently, the performance of mCGT, SVE, and FG% in predicting masking risk was estimated using the AUC.

RESULTS

The AUC values of mCGT and SVE were 0.84 (95% confidence interval, 0.71-0.92) and 0.78 (0.66-0.86), respectively (P = 0.16). The AUC of the FG% was 0.65 (0.52-0.77), which was significantly lower than that of mCGT (P < 0.001). The sensitivity and specificity of mCGT in predicting negative detection were 89% and 71%, respectively; of SVE 83% and 61% (versus 72% and 57% with FG%), suggesting that mCGT was superior to FG% in both sensitivity and specificity, and comparable with SVE.

CONCLUSIONS

Objective mCGT calculated from the volumetric measurement system will highly likely be useful in evaluating breast density and supporting visual assessment for masking risk stratification.

Clinicopathologic Features of Breast Cancer in Egypt-Contemporary Profile and Future Needs: A Systematic Review and Meta-Analysis

PURPOSE

Breast cancer (BC) is the most common cancer among Egyptian females. No current national cancer database is available in Egypt to provide reliable data on the specific clinicopathologic features of BC in this population. Herein, we investigated the clinical profile of BC among Egyptian women.

METHODS

A systematic review of studies on BC published from inception until December 2021 was performed. We explored pooled estimated proportions of different stages of BC at presentation in Egypt and other clinicopathologic features including age, menopausal status, tumor (T) and lymph node (N) stages, and biological subtypes. Data analysis was performed using meta package (R).

RESULTS

Twenty-six studies were eligible for our systematic review and meta-analysis, including 31,172 BC cases. In 12 studies, including 15,067 patients with BC, the estimated mean age was 50.46 years (95% CI, 48.7 to 52.1; I2, 99%), with a pooled proportion of premenopausal/perimenopausal women of 57% (95% CI, 50 to 63; I2, 98%). Among 9,738 patients with BC, pooled proportions of stage I, II, III, and IV were 6% (95% CI, 4 to 8; I2, 90%), 37% (95% CI, 31 to 43; I2, 93%), 45% (95% CI, 42 to 49; I2, 78%), and 11% (95% CI, 9 to 15; I2, 87%), respectively. The pooled proportions of patients with T3 and T4 tumors were 21% (95% CI, 14 to 31; I2, 99%) and 8% (95% CI, 5 to 12; I2, 96%), respectively, while those with positive lymph nodes were 70% (95% CI, 59 to 79; I2, 99%).

CONCLUSION

Dominance of advanced stage and young age at diagnosis represented the two main features of BC among Egyptian women. Our data may serve to guide the policymakers in Egypt as well as other countries with lower resources to prioritize the diagnostic and therapeutic needs in this context.

Test-set training improves the detection rates of invasive cancer in screening mammography

AIM

To investigate if mammographic test-set participation affects routine breast cancer screening performance.

MATERIALS AND METHODS

Clinical audit data between 2008 and 2018 were collected for 35 breast screen readers who participated in the BreastScreen Reader Assessment Strategy (BREAST) and 22 readers with no history of test-set participation. For BREAST readers, the annual audit data were divided according to the year they completed their first test set, and the same years were used randomly to align and divide the data of non-BREAST readers into pre- and post-test set periods. Multiple audit parameters were inspected retrospectively for the two cohorts to identify how their reading performance has evolved in screening mammography.

RESULTS

Investigating 2 calendar years before and after test-set participation, BREAST and non-BREAST readers recalled lower rates of women in the latter period (p=0.03 and p=0.02, respectively). They also improved their positive predictive value (PPV; p=0.01 and p=0.02, respectively). BREAST readers additionally improved their detection rates of invasive cancer (p=0.02) and all cancers (p=0.01). In an extended 3-year comparison, similar improvements occurred in the recall rate for BREAST (p=0.02) and non-BREAST readers (p=0.02) and in PPV (p=0.001, 0.01, respectively); however, improvements in detection rates also occurred exclusively in BREAST readers' performance for invasive cancer (p=0.04), DCIS (p=0.05), and all cancers (p=0.02); however, significant improvements in detection did not involve <15 mm invasive cancers in both periods. Meanwhile, non-BREAST readers demonstrated a decrease in sensitivity (p=0.02).

CONCLUSION

Participation in test sets is linked to over-time improvements in most audit-measured cancer detection rates.

Age-related change in mammographic breast density of women without history of breast cancer over a 10-year retrospective study

Background

Women with higher breast density are at higher risk of developing breast cancer. Breast density is known to affect sensitivity to mammography and to decrease with age. However, the age change and associated factors involved are still unknown. This study aimed to investigate changes in breast density and the associated factors over a 10-year period.

Materials and Methods

The study included 221 women who had undergone eight or more mammograms for 10 years (2011-2020), were between 25 and 65 years of age, and had no abnormalities as of 2011. Breast density on mammographic images was classified into four categories: fatty, scattered, heterogeneously dense, and extremely dense. Breast density was determined using an image classification program with a Microsoft Lobe's machine-learning model. The temporal changes in breast density over a 10-year period were classified into three categories: no change, decrease, and increase. An ordinal logistic analysis was performed with the three groups of temporal changes in breast density categories as the objective variable and the four items of breast density at the start, BMI, age, and changes in BMI as explanatory variables.

Results

As of 2011, the mean age of the 221 patients was 47 ± 7.3 years, and breast density category 3 scattered was the most common (67.0%). The 10-year change in breast density was 64.7% unchanged, 25.3% decreased, and 10% increased. BMI was increased by 64.7% of women. Breast density decreased in 76.6% of the category at the start: extremely dense breast density at the start was correlated with body mass index (BMI). The results of the ordinal logistic analysis indicated that contributing factors to breast density classification were higher breast density at the start (odds ratio = 0.044; 95% CI [0.025-0.076]), higher BMI at the start (odds ratio = 0.76; 95% CI [0.70-0.83]), increased BMI (odds ratio = 0.57; 95% CI [0.36-0.92]), and age in the 40s at the start (odds ratio = 0.49; 95% CI [0.24-0.99]). No statistically significant differences were found for medical history.

Conclusion

Breast density decreased in approximately 25% of women over a 10-year period. Women with decreased breast density tended to have higher breast density or higher BMI at the start. This effect was more pronounced among women in their 40s at the start. Women with these conditions may experience changes in breast density over time. The present study would be useful to consider effective screening mammography based on breast density.

Australian Women's Responses to Breast Density Information: A Content Analysis

Breast density (BD) is an independent risk factor for breast cancer and reduces mammographic sensitivity. This study explored women's responses and intentions if notified that they had dense breasts.

METHODS

Content analysis was used to assess responses from a written questionnaire undertaken in conjunction with focus groups on BD involving 78 Australian women aged 40-74.

RESULTS

Half the women reported that they would feel a little anxious if notified they had dense breasts, while 29.5% would not feel anxious. The most common theme (29.5%) related to anxiety was the psychosocial impact of the possibility of developing cancer, and women believed that being better informed could help with anxiety (26.9%). When asked what they would do if notified of having dense breasts, the most common response was to consult their doctor for information/advice (38.5%), followed by considering supplemental screening (23%). Consequently, when asked directly, 65.4% were interested in undergoing supplemental screening, while others (10.3%) said they "wouldn't worry about it too much".

DISCUSSION

These findings have important implications for health systems with population-based breast screening programs that are currently considering widespread BD notification in terms of the impact on women, health services and primary care.

Machine Learning Approaches with Textural Features to Calculate Breast Density on Mammography

BACKGROUND

breast cancer (BC) is the world's most prevalent cancer in the female population, with 2.3 million new cases diagnosed worldwide in 2020. The great efforts made to set screening campaigns, early detection programs, and increasingly targeted treatments led to significant improvement in patients' survival. The Full-Field Digital Mammograph (FFDM) is considered the gold standard method for the early diagnosis of BC. From several previous studies, it has emerged that breast density (BD) is a risk factor in the development of BC, affecting the periodicity of screening plans present today at an international level.

OBJECTIVE

in this study, the focus is the development of mammographic image processing techniques that allow the extraction of indicators derived from textural patterns of the mammary parenchyma indicative of BD risk factors.

METHODS

a total of 168 patients were enrolled in the internal training and test set while a total of 51 patients were enrolled to compose the external validation cohort. Different Machine Learning (ML) techniques have been employed to classify breasts based on the values of the tissue density. Textural features were extracted only from breast parenchyma with which to train classifiers, thanks to the aid of ML algorithms.

RESULTS

the accuracy of different tested classifiers varied between 74.15% and 93.55%. The best results were reached by a Support Vector Machine (accuracy of 93.55% and a percentage of true positives and negatives equal to TPP = 94.44% and TNP = 92.31%). The best accuracy was not influenced by the choice of the features selection approach. Considering the external validation cohort, the SVM, as the best classifier with the 7 features selected by a wrapper method, showed an accuracy of 0.95, a sensitivity of 0.96, and a specificity of 0.90.

CONCLUSIONS

our preliminary results showed that the Radiomics analysis and ML approach allow us to objectively identify BD.

Overview of Artificial Intelligence in Breast Cancer Medical Imaging

The heavy global burden and mortality of breast cancer emphasize the importance of early diagnosis and treatment. Imaging detection is one of the main tools used in clinical practice for screening, diagnosis, and treatment efficacy evaluation, and can visualize changes in tumor size and texture before and after treatment. The overwhelming number of images, which lead to a heavy workload for radiologists and a sluggish reporting period, suggests the need for computer-aid detection techniques and platform. In addition, complex and changeable image features, heterogeneous quality of images, and inconsistent interpretation by different radiologists and medical institutions constitute the primary difficulties in breast cancer screening and imaging diagnosis. The advancement of imaging-based artificial intelligence (AI)-assisted tumor diagnosis is an ideal strategy for improving imaging diagnosis efficient and accuracy. By learning from image data input and constructing algorithm models, AI is able to recognize, segment, and diagnose tumor lesion automatically, showing promising application prospects. Furthermore, the rapid advancement of "omics" promotes a deeper and more comprehensive recognition of the nature of cancer. The fascinating relationship between tumor image and molecular characteristics has attracted attention to the radiomic and radiogenomics, which allow us to perform analysis and detection on the molecular level with no need for invasive operations. In this review, we integrate the current developments in AI-assisted imaging diagnosis and discuss the advances of AI-based breast cancer precise diagnosis from a clinical point of view. Although AI-assisted imaging breast cancer screening and detection is an emerging field and draws much attention, the clinical application of AI in tumor lesion recognition, segmentation, and diagnosis is still limited to research or in limited patients' cohort. Randomized clinical trials based on large and high-quality cohort are lacking. This review aims to describe the progress of the imaging-based AI application in breast cancer screening and diagnosis for clinicians.