Early Detection Of Breast Cancer Research Articles

Machine Learning Empowered Breast Cancer Diagnosis: Insights from Coimbra Dataset

Aim: The aim of this work is to succinctly communicate the key aspects of a research study on breast cancer. This includes highlighting the global impact and prevalence of breast cancer, emphasizing the challenges of early diagnosis, discussing the potential of technological advancements, and showcasing the role of machine learning algorithms in the context of liver cancer diagnosis. Background: Cancer, notably breast cancer, represents a global health challenge, claiming a significant toll with 12.5% of new cancer cases annually. The prevalence of breast cancer among women worldwide is alarming, resulting in 2.26 million incidents and the unfortunate loss of 685,000 lives. Objective: This article focuses on the critical aspect of early breast cancer diagnosis, acknowledging its heightened difficulty in developing nations compared to developed counterparts. The potential for advancements in technology to serve as a beacon of hope lies in early identification and timely treatment, offering salvation to numerous women and significantly elevating survival chances. Method: In this intricate landscape, machine learning algorithms, particularly in diagnosing liver cancer at its nascent stages, emerge as instrumental tools. The study employs the latest Coimbra dataset, encompassing nine key attributes and a binary classification attribute, with values 1 and 2 signifying benign and malignant cases, respectively. Result: Supervised machine learning algorithms, including Bayes net, multilayer perceptron, IBK, random committee, and random tree, are meticulously applied. Certain models exhibit superior accuracy, precision, recall, and performance, positioning them as promising cornerstones for breast cancer analysis. Conclusion: This structured abstract highlights the urgent need for effective screening and prevention strategies, emphasizing the potential of advanced technology and machine learning algorithms to play a pivotal role in the early detection and analysis of breast cancer, offering hope for improved outcomes and survival rates.

Enhancing awareness of breast self-examination and pap smears for early detection of breast and cervical cancer

Reproductive health problems, especially breast cancer and cervical cancer are the most common cancers diagnosed in women in Indonesia. Currently, several testing methods to detect breast and cervical cancer early, such as breast self-examination (BSE) and pap smears, have been used to reduce cancer risk factors. However, these two methods are still not widely used. Outreach activities have been proven to increase knowledge about the importance of BSE and pap smears in women. This community service program aims to increase knowledge about the causes and symptoms of breast and cervical cancer, as well as early detection using BSE and pap smears. This service activity plan consisted of preparation, implementation, and evaluation stages. The method used consists of delivering interactive material, discussion, and practice. This activity was attended by 38 participants, consisting of women from PKK cadres in the Andir area, Ciroyom District, the majority of whom were over 35 years old. Based on the results of the post-test, this community service activity was successful in increasing participants' knowledge, attitudes, and behavior towards breast and cervical cancer as well as the early detection methods of BSE and pap smears.

Towards Early Breast Cancer Detection: A Deep Learning Approach

Towards Early Breast Cancer Detection: A Deep Learning Approach

Conscious Techniques Affected the Adolescent’s Ability In Early Breast Cancer Detection

Breast cancer is a malignant tumor and grows in the breast tissue resulting in death. Breast cancer prevention by means of breast self-examination. This study aimed to determine the effect of video media on conscious techniques with a combination of demonstrations on the practical ability of adolescents in early detection of breast cancer. The method used a quasi-experimental design, non-equivalent control group design with a population of 279 and a sample of 40 respondents with inclusion criteria of young women, aged 16 years as students of SMA Negeri 1 Tarik Sidoarjo. The data collection measured skills in 2 groups of respondents. The experimental intervention group used videos and demonstrations, while the intervention control group used lectures and leaflets. The intervention was done in 1 day, the posttest was conducted 7 days after the intervention. The results showed that the average value of the pretest posttest of the two groups increased. The results of the statistical test paired T-Test obtained a significant value of p-value = (0.000 < 0.05) meant that there was an effect of providing video media with conscious techniques with a combination of demonstrations on the practical ability of adolescents in early detection of breast cancer. Educators should develop programs to improve skills for early detection of breast cancer, in collaboration with health workers through health promotion so that all students are aware of health.

Advanced deep learning strategies for breast cancer image analysis

One of the leading causes of cancer-related deaths in women is breast cancer (BC). Patients' chances of survival are improved when this ailment is diagnosed in a timely manner and suitable treatments are prescribed. Thus, the chance of survival increases with the early detection of BC. Deep learning neural networks have garnered significant attention in recent years for use in BC screening, identification, and classification. Even though some encouraging results have surfaced, more improvement and confirmation are necessary. In this regard, the creation and comparison of many deep learning approaches for the early identification and classification of BC from mammography pictures is the main focus of our research. In this study, Innovative deep learning methods are developed by freezing the first 40 layers an dropping the 43 layers of MobileNetV2, InceptionV3 and DenseNet121 pretrained models, this approach was reinforced by using Data augmentation techniques. Our suggested methodologies are tested through simulations on MIAS dataset and SA private dataset. The three modified models achieved very high classification accuracies, specifically the modified DenseNet121 model, which reached 99,1% on MIAS dataset and 98,8% on SA dataset.

Economic impact of screening on postdiagnosis work productivity in Japanese women with breast cancer: a life-table modeling approach.

In Japan, biennial mammography screening has been recommended for the early detection of breast cancer (BC) in women aged 40years or above since 2004 by the Ministry of Health, Labor and Welfare. The purpose of this study is to estimate the economic impact of BC screening on work productivity, using a new measure called the productivity-adjusted life-year (PALY). We used a dynamic life table modeling approach to estimate the work productivity of female patients aged 40-64years diagnosed with BC in 2019 over the year of diagnosis and the subsequent 5years. Changes in life-years, PALYs, and gross domestic product (GDP) were assessed by changing the screening detection rate from the current (34.2%) to an ideal (100%) percentage. Each input for modeling was obtained from the most recent public database available. BC patients were estimated to lose 1903 in life-years, 3596 in PALYs, and US$281 million in GDP at the current screening detection rate compared with the ideal detection rate. On the other hand, the following gains are expected when the current screening detection rate was increased to 40-80%; life-years gain; 168-1325, PALYs gain; 317-2503, GDP gain: US$25-196 million. This study has used modeling to show that detecting BC without screening is associated with a lower work productivity and an economic and life-years loss. Encouraging BC screening may be beneficial to maintaining work productivity after diagnosis.

A comparative performance assessment of artificial intelligence based classifiers and optimized feature reduction technique for breast cancer diagnosis

Breast cancer (BC) is a catastrophic global health concern that causes numerous fatalities worldwide. Early detection of breast cancer may mitigate death rates; however, the prevailing diagnostic procedure for the malignancy necessitates numerous multifaceted laboratory tests that must be performed by medical professionals. In this article machine learning, a branch of Artificial Intelligence (AI), has been employed to improve cancer diagnosis, prognoses and survival rates while reducing the vulnerability of humans. Support Vector Machine (SVM), K-nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF) and Grey Wolf Optimizer (GWO) are implemented to prognosticate breast cancer. Comprehensive insights into the efficacy of these approaches for breast cancer prognosis are provided by the performance assessment that is accomplished using the confusion matrix, Receiver Operating Characteristic (ROC) curves and parallel coordinate plots. Both UCI (University of California Irvine) and SEER (Surveillance, Epidemiology and End Results) datasets have been utilized to confirm the investigation's findings and ensure their generalizability across diverse data sources. The results conclusively demonstrate that SVM is the cohort's most accurate classifier. With a stupendous accuracy rate of 99.1 %, the GWO-SVM compares favorably to all other algorithms. Furthermore, feature reduction approaches such as Minimum Redundancy Maximum Relevance (mRMR), ReliefF and Principal Component Analysis (PCA) are utilized. ReliefF has demonstrated exceptional effectiveness with a maximum accuracy of 98.2 %.

Single-cell transcriptome analysis reveals immune microenvironment changes and insights into the transition from DCIS to IDC with associated prognostic genes

BackgroundDuctal carcinoma in situ (DCIS) of the breast is an early stage of breast cancer, and preventing its progression to invasive ductal carcinoma (IDC) is crucial for the early detection and treatment of breast cancer. Although single-cell transcriptome analysis technology has been widely used in breast cancer research, the biological mechanisms underlying the transition from DCIS to IDC remain poorly understood.ResultsWe identified eight cell types through cell annotation, finding significant differences in T cell proportions between DCIS and IDC. Using this as a basis, we performed pseudotime analysis on T cell subpopulations, revealing that differentially expressed genes primarily regulate immune cell migration and modulation. By intersecting WGCNA results of T cells highly correlated with the subtypes and the differentially expressed genes, we identified six key genes: FGFBP2, GNLY, KLRD1, TYROBP, PRF1, and NKG7. Excluding PRF1, the other five genes were significantly associated with overall survival in breast cancer, highlighting their potential as prognostic biomarkers.ConclusionsWe identified immune cells that may play a role in the progression from DCIS to IDC and uncovered five key genes that can serve as prognostic markers for breast cancer. These findings provide insights into the mechanisms underlying the transition from DCIS to IDC, offering valuable perspectives for future research. Additionally, our results contribute to a better understanding of the biological processes involved in breast cancer progression.

Social determinants of breast cancer screening: a multilevel analysis of proximal and distal factors related to the practice of mammography.

Mammography is crucial for early breast cancer detection. In Latin America, Argentina faces a significant breast cancer burden, with varying mammography rates. The social factors influencing mammography practices remain unclear. This study aimed to identify the proximal and distal social determinants of this practice among Argentinean women using a multilevel approach. This nationwide cross-sectional study included 4,924 women aged 50-70 participating in the 2018 National Risk Factor Survey of Argentina. Two-level logistic models were used to estimate measures of association (ORs) between timely mammography practice (within the last 2 years) and selected covariates (sociodemographics, proximal environment, and distal-level variables). The intraclass correlation coefficient (ICC) and proportional change in variance (PCV) were calculated. 62.8% of women underwent timely mammography. Age (OR=0.96; 95%CI 0.94-0.97), health insurance (OR=2.22; 95%CI 1.87-2.63), education (OR=2.1; 95%CI 1.74-2.64), and income (OR=1.56; 95%CI 1.23-1.97) were associated with mammography practice. Women in non-marital (OR=0.61; 95%CI 0.52-0.72) or larger households (OR=0.61; 95%CI 0.51-0.63) were less likely to have timely mammograph; living in a larger city was positively associated (OR=1.28; 95%CI 1.12-1.46). Women in provinces with higher physician density (OR=1.06; 95%CI 1.01-1.11) and lower maternal mortality ratio (OR=0.9; 95%CI 0.87-0.96) had higher chances of timely mammography. The ICC and PCV suggested that the explored healthcare indicators largely explained the macro-contextual effect. Multilevel factors influenced mammography practices in Argentina. The results highlight disparities linked to sociodemographic characteristics and healthcare resources. Efforts to address social inequalities in breast cancer screening must consider multilevel determinants, including in healthcare settings.

B-352 Effects of COVID-19 on Breast Cancer Positivity: A Retrospective and Predictive Study

Abstract Background The COVID-19 pandemic has had a significant impact on the healthcare system, resulting in overall reduction in access to healthcare services, including the suspension of screening programs and postponement of surgeries and medical treatments. The aim of this study was to evaluate the effects of the pandemic on breast cancer positivity from cases received at a laboratory in São Paulo, Brazil. Methods A retrospective analysis of a database from a pathology laboratory in São Paulo, Brazil, was performed during the 2017-2023 period and classified as following: pre-pandemic (January 2017-February 2020), pandemic (March 2020 - April-2022), and post-pandemic (May 2022-December 2023). Cases were classified as negative or positive for malignancies. Positive anatomopathological examinations (breast biopsies and surgical specimens) considered primary, metastatic mesenchymal or endothelial malignancies. The Mann-Kendall test was used to evaluate the trend, and the PACF (Partial Autocorrelation Function) correlogram was used to analyze historical memory for definition of the autoregressive model. The augmented Dickey-Fuller hypothesis test was applied to analyze whether the series is stationary or not, and the first differences test was applied. The ARIMA model was used for the estimates (1,1,1). Statistics were performed using the gretl software and pymannkendall library in the Python language. Results A total of 74,973 samples were evaluated during the 2017-2023 period. In the three-year pre-pandemic period, 32,644 tests were recorded, with an average of 29.34% positive cases. From the analysis of the pre-pandemic time series, there was a positive trend towards an increase in cases (p < 0.0001, z=6.1360, Tau=0.6955). In a non-pandemic scenario, from March 2020 to February 2021, the number of new tests expected would be 11,535; however, in the pandemic period, only 8,904 were observed, a 22.81% reduction on expected cases. According to the historical average positivity rate in the pre-pandemic period, the expected and observed total were, respectively, 3,384 and 2,566 positive cases for malignancy. Conclusions This study reveals a significant decrease of approximately 22.81% in new breast anatomopathological diagnoses during the COVID-19 pandemic in a laboratory in São Paulo, Brazil. For comparison purposes, we identified a 24.17% of missed opportunities on diagnoses of new malignancy-positive cases, compared to the pre-pandemic period. These findings highlight the need for strategies to mitigate the negative effects of the pandemic on the early detection and treatment of breast cancer.

Artificial Intelligence Algorithm for Subclinical Breast Cancer Detection

Early breast cancer detection is associated with lower morbidity and mortality. To examine whether a commercial artificial intelligence (AI) algorithm for breast cancer detection could estimate the development of future cancer. This retrospective cohort study of 116 495 women aged 50 to 69 years with no prior history of breast cancer before they underwent at least 3 consecutive biennial screening examinations used scores from an AI algorithm (INSIGHT MMG, version 1.1.7.2; Lunit Inc; used September 28, 2022, to April 5, 2023) for breast cancer detection and screening data from multiple, consecutive rounds of mammography performed from September 13, 2004, to December 21, 2018, at 9 breast centers in Norway. The statistical analyses were performed from September 2023 to August 2024. Artificial intelligence algorithm score indicating suspicion for the presence of breast cancer. The algorithm provided a continuous cancer detection score for each examination ranging from 0 to 100, with increasing values indicating a higher likelihood of cancer being present on the current mammogram. Maximum AI algorithm score for cancer detection and absolute difference in score among breasts of women developing screening-detected cancer, women with interval cancer, and women who screened negative. The mean (SD) age at the first study round was 58.5 (4.5) years for 1265 women with screening-detected cancer in the third round, 57.4 (4.6) years for 342 women with interval cancer after 3 negative screening rounds, and 56.4 (4.9) years for 116 495 women without breast cancer all 3 screening rounds. The mean (SD) absolute differences in AI scores among breasts of women developing screening-detected cancer were 21.3 (28.1) at the first study round, 30.7 (32.5) at the second study round, and 79.0 (28.9) at the third study round. The mean (SD) differences prior to interval cancer were 19.7 (27.0) at the first study round, 21.0 (27.7) at the second study round, and 34.0 (33.6) at the third study round. The mean (SD) differences among women who did not develop breast cancer were 9.9 (17.5) at the first study round, 9.6 (17.4) at the second study round, and 9.3 (17.3) at the third study round. Areas under the receiver operating characteristic curve for the absolute difference were 0.63 (95% CI, 0.61-0.65) at the first study round, 0.72 (95% CI, 0.71-0.74) at the second study round, and 0.96 (95% CI, 0.95-0.96) at the third study round for screening-detected cancer and 0.64 (95% CI, 0.61-0.67) at the first study round, 0.65 (95% CI, 0.62-0.68) at the second study round, and 0.77 (95% CI, 0.74-0.79) at the third study round for interval cancers. In this retrospective cohort study of women undergoing screening mammography, mean absolute AI scores were higher for breasts developing vs not developing cancer 4 to 6 years before their eventual detection. These findings suggest that commercial AI algorithms developed for breast cancer detection may identify women at high risk of a future breast cancer, offering a pathway for personalized screening approaches that can lead to earlier cancer diagnosis.

Enhanced breast tumor localization with DRA antenna backscattering and GPR algorithm in microwave imaging

With the rapid rate of increasing breast cancer cases across the globe and restrictions of existing early stage detection techniques, microwave imaging based cancer and malignant cell diagnosing methodology is finding its way. Microwave imaging technology has higher accuracy of detection in the variation of tissue properties. In this article, a dielectric resonator based wide band antenna is designed and investigated for microwave imaging (MWI) of early breast cancer detection due to its improved accuracy. The designed antenna has compact size, broad frequency spectrum, high gain and broadside radiation characteristics. To attain the wider bandwidth from (6.5 GHz–12.5 GHz), two rectangular shaped dielectric slabs are used. Wide band microwave frequency band supports the high resolution images. A C-shaped defected ground structure is used to tune the impedance matching. Dielectric resonator based antenna has higher gain and reduces the squints in antenna impedance. Two-asymmetrical dielectric slabs contribute in the excitation of different resonating modes which in turn enhance the bandwidth. It also contributes to excite multiple resonance mode that in turn enhance the bandwidth. Antenna performance is first numerically and experimentally verified in free space. Then, simulation based performance is examined for the microwave imaging. Numerical phantom with equivalent tissue properties is designed with and without the tumor. Unhealthy cells have higher water percentage and larger dielectric constant as compared to healthy cells. It causes the different in the back scattered signal of antenna, which is analyzed to diagnose the tumor or cancer. Here, a maximum deviation of 16 dB is observed in the backscattered signal. Furthermore, surface of the breast tissue is scanned along the x-axis and y-axis at different angels. The collected signals are used to reconstruct the microwave image of the interior. Mean value data of backscattered signals is used to locate the existence of unhealthy cells and GPR algorithm is used to calculate the depth of a 4 mm tumor in breast tissue. Proposed structure has shown efficient performance for microwave imaging.

Exploring factors associated with uptake of breast cancer screening among a subset of women in a South African metropolitan area: research to inform public health interventions

Aim Breast cancer represents a significant global health concern and remains a leading cause of cancer-related mortality, particularly among women in developing countries. Despite the South African Department of Health recommending early breast cancer detection to improve overall survival rates and enhance patients’ quality of life, there is lack of awareness among women regarding the importance of regular breast cancer screening and available options. This study investigates the factors associated with breast cancer screening uptake among a subset of women attending public health services in a South African metropolitan area. Methods A cross-sectional survey of adult women attending primary health care services at a community health centre was conducted in February-March 2020. Data on sociodemographics, health beliefs, access to information, and breast cancer screening uptake were subjected to descriptive and logistic regression analyses. Results Out of 252 women, 48.0% were older than 30 years, 69.4% were unemployed, and 56.0% self-reported breast cancer screening non-uptake. After adjusting for other variables, marital status, employment status, home language, and perceived breast cancer screening benefits were significantly associated with breast cancer screening uptake. Married women were twice (AoR: 2.3; CI: 1.19-4.58) as likely to self-report breast cancer screening than their unmarried counterparts. Employed women were almost thrice (AoR: 2.6; CI: 1.41-4.83) as likely to self-report breast cancer screening uptake relative to unemployed women. Women whose home language was Sesotho were almost twice (AoR: 1.9; CI: 1.05-3.43) as likely to self-report breast cancer screening than women reporting other home languages. Self-reported uptake of breast cancer screening was also more likely with every unit increase in perceived benefit scores (AoR: 1.1; CI: 1.02-1.16). Conclusions A significant proportion of women were in the age range where breast cancer screening is typically recommended, with more than half self-reporting breast cancer screening non-uptake. Results highlight the necessity of addressing socioeconomic, cultural and individual perceptions to improve breast cancer screening rates and promote early detection.

Recognition and comprehension of breast awareness among hospital staff: a questionnaire survey using the explanatory leaflet in Japan

BackgroundThis study aimed to assess the recognition and understanding of breast awareness (BA) among hospital staff, a group considered influential in disseminating information about health. Compared to the traditional approach of breast self-examination (BSE), BA has gained prominence as a concept focused on early detection. The study also explored the effectiveness of an informational leaflet in conveying BA concepts.MethodsWe conducted an online, voluntary, and anonymous questionnaire survey at St. Luke’s International Hospital in Japan, where approximately 1,000 breast cancer surgeries are performed annually. The survey comprised three sections: pre-leaflet questions, the informational leaflet, and post-leaflet questions.ResultsFrom a pool of 500 completed questionnaires, 499 were deemed suitable for the analysis. Notably, 78% of respondents were unfamiliar with “BA” before the survey. However, 89.1% expressed interest in adopting daily practices for early breast cancer detection. Following the leaflet exposure, 98.4% of respondents claimed to have understood BA, either completely or partially. The leaflet aided 93.2% of these individuals in differentiating between BA and the traditional BSE method. These outcomes remained consistent across various demographic segments such as occupation, age, and experience with breast cancer care.ConclusionsThe study underscores a concerning lack of awareness regarding BA among hospital staff within the surveyed institution. This highlights the need to engage medical professionals in promoting BA within the community. The informational leaflet proved effective in enhancing comprehension of BA across diverse groups, indicating its potential as a widely applicable educational tool. The leaflet facilitated the comprehension of BA among respondents across all demographic groups, indicating its potential for widespread utility.

Dielectric Signatures of Late Carcinoma Immune Cells Using MMTV-PyMT Mammary Carcinoma Models.

Peripheral blood mononuclear cells (PBMCs) are specialized immune cells produced from hematopoietic stem cells (HSC). They actively surveil for any signs of infection, foreign invaders, and abnormal or aberrant cells associated with diseases. Numerous inherent interactions between PBMCs and proliferating cancer cells facilitate cellular communication, inducing alterations in the composition of the PBMCs. These subtle alterations can be detected by using dielectrophoresis (DEP). The ultimate objective is to apply this knowledge in a clinical setting to achieve noninvasive early detection of breast cancer while minimizing the occurrence of false positives and negatives commonly associated with standard screening methods like mammography. To realize our long-term goal, we are probing the dielectric properties of the PBMCs from FVB/N MMTV-PyMT+ (late carcinoma, PyMT+ PBMC) and FVB/N (wild-type, WT-PBMC) age-matched mice at 14+ weeks using dielectrophoresis on a microfluidic platform. The central hypothesis of this research is that the changes triggered in the subcellular components, such as the cytoskeleton, lipid bilayer membrane, cytoplasm, focal adhesion proteins, and extracellular matrix (ECM) at the onset of carcinoma, regulate dielectric properties (conductivity, σ; and permittivity, ε), thus affecting the bioelectric signals that aid in the detection of breast cancer. The ANOVA results suggest a significant difference in PyMT+ PBMCs crossover frequencies at 0.01 and 0.05 S/m medium conductivity levels. Post hoc pairwise analysis of WT-PBMCs showed that the crossover frequencies are distinct across the medium conductivity ranges from 0.01 to 0.05 S/m. This study revealed that on average, PyMT+ PBMCs have increased crossover frequency, polarizability, higher membrane capacitance, and a folding factor compared with the age-matched wild-type PBMCs.

An Artificial Intelligence-Based Tool for Enhancing Pectoral Muscle Segmentation in Mammograms: Addressing Class Imbalance and Validation Challenges in Automated Breast Cancer Diagnosis.

Breast cancer remains a major health concern worldwide, requiring the advancement of early detection methods to improve prognosis and treatment outcomes. In this sense, mammography is regarded as the gold standard in breast cancer screening and early detection. However, in a scenario where extensive analysis is required, a large set of mammograms conducted by radiologists may carry out false negative or false positive diagnoses. Therefore, artificial intelligence has emerged in recent years as a method for enhancing timing in breast cancer diagnosis. Nonetheless, preprocessing stages are required to prepare the mammography dataset to enhance learning models to correctly identify breast anomalies. In this paper, we introduce a novel method employing convolutional neural networks (CNNs) to segment the pectoral muscle in 1288 mediolateral oblique mammograms (MLOs), thereby addressing class imbalance and overfitting between classes, and dataset augmentation based on translation, rotation, and scale transformation. The effectiveness of the model was assessed through a confusion matrix and performance metrics, highlighting an average Dice coefficient of 0.98 and a Jaccard index of 0.96. The outcomes demonstrate the model capability to accurately identify three classes: pectoral muscle, breast, and background. This study emphasizes the importance of tackling class imbalance problems and augmenting data for the training of models for reliable early breast cancer detection.

Hybrid ensemble deep learning model for advancing breast cancer detection and classification in clinical applications

Being the most common type of cancer worldwide, and affecting over 2.3 million women, breast cancer poses a significant health threat. Although survival rates have improved around the world due to advances in screening, diagnosis, and treatment, early detection remains crucial for effective management. This study seeks to introduce a novel hybrid model that makes use of image-preprocessing techniques and deep-learning algorithms on mammograms to enhance the detection and classification accuracy of breast cancer lesions. The model was tested on a dataset comprising 20,000 mammograms. First, image-processing techniques, such as Contrast-Limited Adaptive Histogram Equalization, Gaussian Blur, and sharpening methods were used to optimize the images for enhanced feature extraction. In addition, the Ensemble Deep Random Vector-Functional Link Neural Network algorithm, YOLOv5, and MedSAM segmentation models were utilized for robust deep learning-based extraction, classification, and visualization of lesions. Finally, the model was clinically validated on 800 patients. The study found a notable enhancement in both accuracy and processing time for benign and malignant diagnoses using the hybrid model. The model achieves an impressive accuracy of 99.7 % and demonstrates a remarkable processing time of 0.75 s. In clinical applications, the hybrid model exhibits high proficiency, reporting 97.2 % accuracy for benign cases and 98.6 % for malignant scenarios. These results highlight the effectiveness of the hybrid model in improving diagnostic accuracy, offering a promising tool for early breast cancer detection.

Abstract A119: Disparities in mammography screening adherence throughout phases of the COVID-19 Pandemic: Results from the 2018, 2020, and 2022 Behavioral Risk Factor Surveillance System

Abstract Background Early detection of breast cancer through mammography screening reduces mortality risk. Disruptions during the COVID-19 pandemic in the United States worsened disparities in mammography use, and it is unclear if the impact persisted after the rollout of COVID-19 vaccine in 2021. We aim to describe disparities among minority and underserved women in breast cancer screening before and during the initial phase of the COVID-19 pandemic and assess any changes in screening patterns following the rollout of COVID-19 vaccines in the United States. Methods We used Behavioral Risk Factor Surveillance System (BRFSS) data for years before (2018), during (2020), and after (2022) the COVID-19 vaccine rollout in the United States. We estimated the proportion of women aged 50–74 years who were non-adherent to the United States Preventive Services Taskforce screening guidelines (no mammography in the past two years) by select sociodemographic and behavioral factors. Weighted frequencies, Chi-square tests with Rao-Scott correction, and log-binomial regression models with a Poisson family were used to estimate prevalence ratios (PR) for non-adherence by selected factors to identify sub-populations more likely to be non-adherent. Results A total of 116,756, 102,778, and 110,456 women were included in the analytic samples from 2018, 2020, and 2022, respectively. The prevalence of non-adherence to breast cancer screening guidelines was 21.1%, 20.6%, and 21.6%, in 2018, 2020, and 2022, respectively. Comparing 2018 to 2022, the proportion of non-adherent women significantly increased among those who did not graduate high school (26.7% vs 34.7%), those without health insurance (45.3% vs 61.9%), and women with incomes below $15,000 (28.1% vs 35.1%). Insurance status had the largest impact on screening adherence, with uninsured women more than twice as likely to be non-adherent compared to insured women both before (PR = 2.15, 95% CI: 1.94, 2.38) and after the COVID-19 vaccine rollout (2.24, 95% CI: 2.05, 2.45). Prevalence ratios of nonadherence significantly increased between 2018 and 2020, then returned back to pre-pandemic levels in 2022 among those living in non-metropolitan versus metropolitan areas, never married versus married, self-employed versus employed, and those reporting poor versus excellent health. Hispanics were more likely than non-Hispanics to be non-adherent in 2020 and 2022, but no difference was seen in 2018. Conclusions Persistent disparities in breast cancer screening adherence among eligible women in the United States were observed throughout the phases of the COVID-19 pandemic. Although overall adherence remained stable, certain subgroups, such as those with lower education, no health insurance, Hispanic, rural residence, and lower income, experienced worsening adherence. Health insurance status was the most significant factor associated with non-adherence, highlighting the need for targeted interventions among those who are uninsured. Citation Format: Gregory Chang C Chang, James L. Fisher, Electra D. Paskett. Disparities in mammography screening adherence throughout phases of the COVID-19 Pandemic: Results from the 2018, 2020, and 2022 Behavioral Risk Factor Surveillance System [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr A119.

Abstract A154: Mammograms Across Missouri: A Qualitative Evaluation among Healthcare Providers to Improve Breast Cancer Screening in Rural Communities

Abstract Introduction: Breast cancer is the leading cause of cancer-related deaths among women, with rural communities having the highest rates of breast cancer mortality. Despite this burden, many women begin mammography later than recommended and the mammography screening rate remains low, with a previously reported declining trend, particularly among rural women compared to urban women. This study seeks to identify barriers to breast cancer screening and examine strategies to overcome these barriers in rural Missouri. Methods: In partnership with the Missouri Cancer Consortium and the Missouri Comprehensive Cancer Control Programs, we conducted key informant interviews with mammography health providers at health centers in each of ten counties with the highest breast cancer mortality and lowest screening mammography rates in Missouri. Semi-structured interviews included questions about seeking and delivering breast cancer screening as well as provider experiences with patient navigation, mobile mammography, and the CDC’s National Breast and Cervical Cancer Early Detection Program (NBCCEDP). Data were analyzed using a deductive-inductive thematic approach in NVivo 14. Results: We conducted 14 interviews with health providers across 12 health sites. The majority of these sites used patient navigation (67%), mobile mammography (80%), and the CDC NBCCEDP (67%) as strategies to improve access to breast healthcare. Health providers frequently mentioned “no-shows” as a major challenge to providing breast healthcare, along with limited transportation, cost, fear, and competing work priorities as the major barriers faced by patients in seeking breast healthcare. Patient navigation and mobile mammography were beneficial in bringing breast healthcare closer to the community, while the CDC NBCCEDP covered screening and follow-up costs. The main facilitating factors to implementing these strategies were having dedicated staff and funding for patient navigation, and existing partnerships with other health institutions for mobile mammography. However, health providers identified several implementing challenges for the strategies. This included limited staff capacity and difficulty contacting or following up patients to schedule appointments for patient navigation and mobile mammography. For the CDC NBCCEDP, the challenges included difficulty identifying eligible patients and extensive paperwork in the enrollment process. Providers recommended increasing staff capacity to support patient navigation, using digital technology in identifying and enrolling women for the CDC NBCCEDP, and integrating mobile mammography with the CDC NBCCEDP to improve implementation efficiency. Discussion: The integration of multiple component strategies within breast healthcare may address the multi-level barriers women face in accessing breast cancer screening. This study contributes to the identification of organizational supports and informs development of policies needed to enhance the adoption and implementation of strategies that support the effective delivery of breast healthcare. Citation Format: Thembekile Shato, Carmen Vardeman, Sarah J. Chavez, Ashley C. Holmes, Joann Lee, Xarria Lewis, Katie Manga, Misty Phillips, Keisha Windham, Bettina F. Drake. Mammograms Across Missouri: A Qualitative Evaluation among Healthcare Providers to Improve Breast Cancer Screening in Rural Communities [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr A154.

Digital Breast Tomosynthesis Screening Improves Early Breast Cancer Detection and Survival in Taiwan.

Our objective was to compare the efficacy of digital breast tomosynthesis (DBT) and digital mammography (DM) in breast cancer screening and their impact on long-term overall survival (OS). The study involved 48 549 consecutive mammography examinations between 2011 and 2015 at a medical center in Taiwan, identifying 545 women who were screened and diagnosed with breast cancer. Digital mammography and DBT examinations were alternated on different days. Patients were categorized based on mammographic modality, breast density, and American Joint Committee on Cancer (AJCC) stage. To determine the long-term outcome until August 2021, survival rates were analyzed using the Kaplan-Meier (K-M) survival analysis. The mean age at breast cancer diagnosis was 53.2 years. Digital breast tomosynthesis examinations were significantly associated with early breast cancer (AJCC stage 0 to 2) (P = .022). The 5- and 9-year OS rates for all patients were 96.8% and 93.0%, respectively. The 5- and 9-year OS was significantly greater in the DBT group (98.4% and 96.8%) compared with the DM group (95.0% and 90.4%) (P = .030 for all). The K-M survival analysis demonstrated a significantly higher OS in the DBT group than the DM group (P = .037). Furthermore, DBT significantly improved OS in a cohort of women with stage II and III cancer (P = .032) and heterogeneously dense breasts (P = .045). Screening with DBT is associated with early breast cancer diagnosis and higher survival rates compared with DM.