Phantom-based analysis of variations in automatic exposure control across three mammography systems: implications for radiation dose and image quality in mammography, DBT, and CEM.

BACKGROUND: Automatic exposure control (AEC) plays a crucial role in mammography by determining the exposure conditions needed to achieve specific image quality based on the absorption characteristics of compressed breasts. This study aimed to characterize the behavior of AEC for digital mammography (DM), digital breast tomosynthesis (DBT), and low-energy (LE) and high-energy (HE) acquisitions used in contrast-enhanced mammography (CEM) for three mammography systems from two manufacturers. METHODS: Using phantoms simulating various breast thicknesses, 363 studies were acquired using all available AEC modes 165 DM, 132 DBT, and 66 LE-CEM and HE-CEM. AEC behaviors were compared across systems and modalities to assess the impact of different technical components and manufacturers' strategies on the resulting mean glandular doses (MGDs) and image quality metrics such as contrast-to-noise ratio (CNR). RESULTS: For all systems and modalities, AEC increased MGD for increasing phantom thicknesses and decreased CNR. The median MGD values (interquartile ranges) were 1.135 mGy (0.772-1.668) for DM, 1.257 mGy (0.971-1.863) for DBT, 1.280 mGy (0.937-1.878) for LE-CEM, and 0.630 mGy (0.397-0.713) for HE-CEM. Medians CNRs were 14.2 (7.8-20.2) for DM, 4.91 (2.58-7.20) for a single projection in DBT, 11.9 (8.0-18.2) for LE-CEM, and 5.2 (3.6-9.2) for HE-CEM. AECs showed high repeatability, with variations lower than 5% for all modes in DM, DBT, and CEM. CONCLUSIONS: The study revealed substantial differences in AEC behavior between systems, modalities, and AEC modes, influenced by technical components and manufacturers' strategies, with potential implications in radiation dose and image quality in clinical settings. RELEVANCE STATEMENT: The study emphasized the central role of automatic exposure control in DM, DBT, and CEM acquisitions and the great variability in dose and image quality among manufacturers and between modalities. Caution is needed when generalizing conclusions about differences across mammography modalities. KEY POINTS: • AEC plays a crucial role in DM, DBT, and CEM. • AEC determines the "optimal" exposure conditions needed to achieve specific image quality. • The study revealed substantial differences in AEC behavior, influenced by differences in technical components and strategies.

MAM-E: Mammographic Synthetic Image Generation with Diffusion Models.

Generative models are used as an alternative data augmentation technique to alleviate the data scarcity problem faced in the medical imaging field. Diffusion models have gathered special attention due to their innovative generation approach, the high quality of the generated images, and their relatively less complex training process compared with Generative Adversarial Networks. Still, the implementation of such models in the medical domain remains at an early stage. In this work, we propose exploring the use of diffusion models for the generation of high-quality, full-field digital mammograms using state-of-the-art conditional diffusion pipelines. Additionally, we propose using stable diffusion models for the inpainting of synthetic mass-like lesions on healthy mammograms. We introduce MAM-E, a pipeline of generative models for high-quality mammography synthesis controlled by a text prompt and capable of generating synthetic mass-like lesions on specific regions of the breast. Finally, we provide quantitative and qualitative assessment of the generated images and easy-to-use graphical user interfaces for mammography synthesis.

Assessing breast arterial calcification in mammograms and its implications for atherosclerotic cardiovascular disease risk.

PURPOSE: Breast arterial calcifications (BAC) are incidentally observed on mammograms, yet their implications remain unclear. We investigated lifestyle, reproductive, and cardiovascular determinants of BAC in women undergoing mammography screening. Further, we investigated the relationship between BAC, coronary arterial calcifications (CAC) and estimated 10-year atherosclerotic cardiovascular (ASCVD) risk. METHODS: In this cross-sectional study, we obtained reproductive history and CVD risk factors from 215 women aged 18 or older who underwent mammography and cardiac computed tomographic angiography (CCTA) within a 2-year period between 2007 and 2017 at hospital. BAC was categorized as binary (present/absent) and semi-quantitatively (mild, moderate, severe). CAC was determined using the Agatston method and recorded as binary (present/absent). Adjusted odds ratios (ORs) and 95 % confidence intervals (CIs) were calculated, accounting for age as a confounding factor. ASCVD risk over a 10-year period was calculated using the Pooled Cohort Risk Equations. RESULTS: Older age, systolic and diastolic blood pressures, higher parity, and younger age at first birth (≤28 years) were significantly associated with greater odds of BAC. Women with both BAC and CAC had the highest estimated 10-year risk of ASCVD (13.30 %). Those with only BAC (8.80 %), only CAC (5.80 %), and no BAC or CAC (4.40 %) had lower estimated 10-year risks of ASCVD. No association was detected between presence of BAC and CAC. CONCLUSIONS: These findings support the hypothesis that BAC on a screening mammogram may help to identify women at potentially increased risk of future cardiovascular disease without additional cost and radiation exposure.

Evaluation of clinical profiles, imaging findings and antituberculosis treatment outcome in granulomatous mastitis: An Indian scenario.

BACKGROUND: The management of choice for granulomatous mastitis (GM) has yet to be determined but few studies have demonstrated that anti-tubercular treatment (ATT) could be an effective alternative therapeutic option. Hence, the objective of the current study is to determine the clinical feature, radiological imaging findings, and histopathological examination results exhibited by GM and tuberculosis (TB)-proven GM as well as to evaluate the ATT clinical outcome in GM patients. METHODS: The study was performed on 68 GM patients who were referred to the department of pulmonology by the breast clinic (from January 2018 to August 2021). Study populations were categorized into two groups GM and TB-proven GM patients and all were prescribed with standard ATT regimen and were continuously followed up. SPSS version 25 was employed for statistical assessment. RESULTS: Our study showed that 6 patients from GM and 4 patients from the TB-proven GM group got relapsed. For patients who displayed partial remission, ATT treatment was started after assessing the side effects potential. 14.6% (n = 6) and 7.4% (n = 2) patients who initially demonstrated partial remission were also completely cured. ATT treatment curable rate was determined to be 90% (n = 37) and 81.5% (n = 22) for GM and TB-proven GM patients correspondingly. Therefore, the current study demonstrated nil significant differences between groups. CONCLUSION: The current study warrants that ATT therapy could be an effective and better treatment of choice for GM patients irrespective of their clinical condition.

Deep transfer learning with fuzzy ensemble approach for the early detection of breast cancer.

Breast Cancer is a significant global health challenge, particularly affecting women with higher mortality compared with other cancer types. Timely detection of such cancer types is crucial, and recent research, employing deep learning techniques, shows promise in earlier detection. The research focuses on the early detection of such tumors using mammogram images with deep-learning models. The paper utilized four public databases where a similar amount of 986 mammograms each for three classes (normal, benign, malignant) are taken for evaluation. Herein, three deep CNN models such as VGG-11, Inception v3, and ResNet50 are employed as base classifiers. The research adopts an ensemble method where the proposed approach makes use of the modified Gompertz function for building a fuzzy ranking of the base classification models and their decision scores are integrated in an adaptive manner for constructing the final prediction of results. The classification results of the proposed fuzzy ensemble approach outperform transfer learning models and other ensemble approaches such as weighted average and Sugeno integral techniques. The proposed ResNet50 ensemble network using the modified Gompertz function-based fuzzy ranking approach provides a superior classification accuracy of 98.986%.

Use of an AI Score Combining Cancer Signs, Masking, and Risk to Select Patients for Supplemental Breast Cancer Screening.

Background Mammographic density measurements are used to identify patients who should undergo supplemental imaging for breast cancer detection, but artificial intelligence (AI) image analysis may be more effective. Purpose To assess whether AISmartDensity-an AI-based score integrating cancer signs, masking, and risk-surpasses measurements of mammographic density in identifying patients for supplemental breast imaging after a negative screening mammogram. Materials and Methods This retrospective study included randomly selected individuals who underwent screening mammography at Karolinska University Hospital between January 2008 and December 2015. The models in AISmartDensity were trained and validated using nonoverlapping data. The ability of AISmartDensity to identify future cancer in patients with a negative screening mammogram was evaluated and compared with that of mammographic density models. Sensitivity and positive predictive value (PPV) were calculated for the top 8% of scores, mimicking the proportion of patients in the Breast Imaging Reporting and Data System "extremely dense" category. Model performance was evaluated using area under the receiver operating characteristic curve (AUC) and was compared using the DeLong test. Results The study population included 65 325 examinations (median patient age, 53 years [IQR, 47-62 years])-64 870 examinations in healthy patients and 455 examinations in patients with breast cancer diagnosed within 3 years of a negative screening mammogram. The AUC for detecting subsequent cancers was 0.72 and 0.61 (P < .001) for AISmartDensity and the best-performing density model (age-adjusted dense area), respectively. For examinations with scores in the top 8%, AISmartDensity identified 152 of 455 (33%) future cancers with a PPV of 2.91%, whereas the best-performing density model (age-adjusted dense area) identified 57 of 455 (13%) future cancers with a PPV of 1.09% (P < .001). AISmartDensity identified 32% (41 of 130) and 34% (111 of 325) of interval and next-round screen-detected cancers, whereas the best-performing density model (dense area) identified 16% (21 of 130) and 9% (30 of 325), respectively. Conclusion AISmartDensity, integrating cancer signs, masking, and risk, outperformed traditional density models in identifying patients for supplemental imaging after a negative screening mammogram. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Kim and Chang in this issue.

Investigating the detection of breast cancer with deep transfer learning using ResNet18 and ResNet34.

A lot of underdeveloped nations particularly in Africa struggle with cancer-related, deadly diseases. Particularly in women, the incidence of breast cancer is rising daily because of ignorance and delayed diagnosis. Only by correctly identifying and diagnosing cancer in its very early stages of development can be effectively treated. The classification of cancer can be accelerated and automated with the aid of computer-aided diagnosis and medical image analysis techniques. This research provides the use of transfer learning from a Residual Network 18 (ResNet18) and Residual Network 34 (ResNet34) architectures to detect breast cancer. The study examined how breast cancer can be identified in breast mammography pictures using transfer learning from ResNet18 and ResNet34, and developed a demo app for radiologists using the trained models with the best validation accuracy. 1, 200 datasets of breast x-ray mammography images from the National Radiological Society's (NRS) archives were employed in the study. The dataset was categorised as implant cancer negative, implant cancer positive, cancer negative and cancer positive in order to increase the consistency of x-ray mammography images classification and produce better features. For the multi-class classification of the images, the study gave an average accuracy for binary classification of benign or malignant cancer cases of 86.7% validation accuracy for ResNet34 and 92% validation accuracy for ResNet18. A prototype web application showcasing ResNet18 performance has been created. The acquired results show how transfer learning can improve the accuracy of breast cancer detection, providing invaluable assistance to medical professionals, particularly in an African scenario.

Imaging for local recurrence of breast cancer.

PURPOSE: Isolated locoregional recurrence of breast cancer (ILRR) and contralateral breast cancer (CBC) affect up to 20% of all breast cancer (BC) patients in the first 20 years after primary diagnosis. Treatment options comprise surgical interventions and further systemic therapies depending on the histological subtype. Patients with hereditary breast or ovarian cancer (HBOC) undergo MRI, mammography, and ultrasound in the aftercare of BC, while non-HBOC (nHBOC) patients do not regularly receive MRI. Since early detection is crucial for morbidity and mortality, the evaluation and constant improvement of imaging methods of the breast is necessary. METHODS: We retrospectively analyzed the data of 1499 former BC patients that received imaging of the breast at a tertiary-care university hospital between 2015 and 2020. The analysis comprised various patient characteristics, such as breast density, age, tumor size and subtype, and their influence on BC detection rates by the different imaging methods. RESULTS: Within the patient sample, 176 individuals (11.7% of former BC patients) were diagnosed with either ILRR or CBC. CBC was observed in 32.4% of patients, while both ILRR and secondary breast cancer occurred in 20.5% and 23.9% of all patients. Sensitivity of MRI, mammography, and ultrasound for recurrent malignancy was 97.9%, 66.3%, and 67.8%, respectively. ILRR and CBC detection rates were similar for patients with and without HBOC history. Lower breast density and larger tumor size increased the detection rates of all imaging modalities. CONCLUSION: In breast cancer survivors, MRI might improve the early detection of ILRR and CBC in both HBOC and nHBOC patients.

Ultrasonography is an effective tool for breast cancer screening in individuals with severe motor and intellectual disabilities.

BACKGROUND: Individuals with severe motor and intellectual disabilities have become an aging population, and high cancer morbidity and mortality are critical issues affecting their survival. Cancer screening is a crucial method of resolving this issue; however, a suitable screening method for them has not been established. METHODS: We used ultrasonography alone and performed breast cancer screening for women over 30 years old in our facility from 2016 to 2023. We observed the outcomes and calculated the recall/detection rate in this screening. RESULTS: Three cases among 379 tested positive in this screening, all of which underwent radical surgery. They are alive and well without relapse present. We detected these breast cancer cases with a low recall rate. CONCLUSION: We were able to successfully detect breast cancer cases using ultrasonography alone. Ultrasonography is an effective and feasible tool for breast cancer screening in individuals with severe motor and intellectual disabilities.

Application of deep learning on mammographies to discriminate between low and high-risk DCIS for patient participation in active surveillance trials.

BACKGROUND: Ductal Carcinoma In Situ (DCIS) can progress to invasive breast cancer, but most DCIS lesions never will. Therefore, four clinical trials (COMET, LORIS, LORETTA, AND LORD) test whether active surveillance for women with low-risk Ductal carcinoma In Situ is safe (E. S. Hwang et al., BMJ Open, 9: e026797, 2019, A. Francis et al., Eur J Cancer. 51: 2296-2303, 2015, Chizuko Kanbayashi et al. The international collaboration of active surveillance trials for low-risk DCIS (LORIS, LORD, COMET, LORETTA),  L. E. Elshof et al., Eur J Cancer, 51, 1497-510, 2015). Low-risk is defined as grade I or II DCIS. Because DCIS grade is a major eligibility criteria in these trials, it would be very helpful to assess DCIS grade on mammography, informed by grade assessed on DCIS histopathology in pre-surgery biopsies, since surgery will not be performed on a significant number of patients participating in these trials. OBJECTIVE: To assess the performance and clinical utility of a convolutional neural network (CNN) in discriminating high-risk (grade III) DCIS and/or Invasive Breast Cancer (IBC) from low-risk (grade I/II) DCIS based on mammographic features. We explored whether the CNN could be used as a decision support tool, from excluding high-risk patients for active surveillance. METHODS: In this single centre retrospective study, 464 patients diagnosed with DCIS based on pre-surgery biopsy between 2000 and 2014 were included. The collection of mammography images was partitioned on a patient-level into two subsets, one for training containing 80% of cases (371 cases, 681 images) and 20% (93 cases, 173 images) for testing. A deep learning model based on the U-Net CNN was trained and validated on 681 two-dimensional mammograms. Classification performance was assessed with the Area Under the Curve (AUC) receiver operating characteristic and predictive values on the test set for predicting high risk DCIS-and high-risk DCIS and/ or IBC from low-risk DCIS. RESULTS: When classifying DCIS as high-risk, the deep learning network achieved a Positive Predictive Value (PPV) of 0.40, Negative Predictive Value (NPV) of 0.91 and an AUC of 0.72 on the test dataset. For distinguishing high-risk and/or upstaged DCIS (occult invasive breast cancer) from low-risk DCIS a PPV of 0.80, a NPV of 0.84 and an AUC of 0.76 were achieved. CONCLUSION: For both scenarios (DCIS grade I/II vs. III, DCIS grade I/II vs. III and/or IBC) AUCs were high, 0.72 and 0.76, respectively, concluding that our convolutional neural network can discriminate low-grade from high-grade DCIS.

Long-term Mammography Screening Trends and Predictors of Return to Screening after the COVID-19 Pandemic: Results from a Statewide Registry.

Purpose To evaluate long-term trends in mammography screening rates and identify sociodemographic and breast cancer risk characteristics associated with return to screening after the COVID-19 pandemic. Materials and Methods In this retrospective study, statewide screening mammography data of 222 384 female individuals aged 40 years or older (mean age, 58.8 years ± 11.7 [SD]) from the Vermont Breast Cancer Surveillance System were evaluated to generate descriptive statistics and Joinpoint models to characterize screening patterns during 2000-2022. Log-binomial regression models estimated associations of sociodemographic and risk characteristics with post-COVID-19 pandemic return to screening. Results The proportion of female individuals in Vermont aged 50-74 years with a screening mammogram obtained in the previous 2 years declined from a prepandemic level of 61.3% (95% CI: 61.1%, 61.6%) in 2019 to 56.0% (95% CI: 55.7%, 56.3%) in 2021 before rebounding to 60.7% (95% CI: 60.4%, 61.0%) in 2022. Screening adherence in 2022 remained substantially lower than that observed during the 2007-2010 apex of screening adherence (66.1%-67.0%). Joinpoint models estimated an annual percent change of -1.1% (95% CI: -1.5%, -0.8%) during 2010-2022. Among the cohort of 95 644 individuals screened during January 2018-March 2020, the probability of returning to screening during 2020-2022 varied by age (eg, risk ratio [RR] = 0.94 [95% CI: 0.93, 0.95] for age 40-44 vs age 60-64 years), race and ethnicity (RR = 0.84 [95% CI: 0.78, 0.90] for Black vs White individuals), education (RR = 0.84 [95% CI: 0.81, 0.86] for less than high school degree vs college degree), and by 5-year breast cancer risk (RR = 1.06 [95% CI: 1.04, 1.08] for very high vs average risk). Conclusion Despite a rebound to near prepandemic levels, Vermont mammography screening rates have steadily declined since 2010, with certain sociodemographic groups less likely to return to screening after the pandemic. Keywords: Mammography, Breast, Health Policy and Practice, Neoplasms-Primary, Epidemiology, Screening Supplemental material is available for this article. © RSNA, 2024.

Healthcare professionals' experiences of caring for women with false-positive screening test results in the National Health Service Breast Screening Programme.

BACKGROUND: Understanding healthcare professionals' (HCPs) experiences of caring for women with false-positive screening test results in the National Health Service Breast Screening Programme (NHSBSP) is important for reducing the impact of such results. METHODS: Interviews were undertaken with 12 HCPs from a single NHSBSP unit, including advanced radiographer practitioners, breast radiographers, breast radiologists, clinical nurse specialists (CNSs), and a radiology healthcare assistant. Data were analysed thematically using Template Analysis. RESULTS: Two themes were produced: (1) Gauging and navigating women's anxiety during screening assessment was an inevitable and necessary task for all participants. CNSs were perceived as particularly adept at this, while breast radiographers reported a lack of adequate formal training. (2) Controlling the delivery of information to women (including amount, type and timing of information). HCPs reported various communication strategies to facilitate women's information processing and retention during a distressing time. CONCLUSIONS: Women's anxiety could be reduced through dedicated CNS support, but this should not replace support from other HCPs. Breast radiographers may benefit from more training to emotionally support recalled women. While HCPs emphasised taking a patient-centred communication approach, the use of other strategies (e.g., standardised scripts) and the constraints of the 'one-stop shop' model pose challenges to such an approach. PATIENT AND PUBLIC CONTRIBUTION: During the study design, two Patient and Public Involvement members (women with false-positive-breast screening test results) were consulted to gain an understanding of patient perspectives and experiences of being recalled specifically in the NHSBSP. Their feedback informed the formulations of the research aim, objectives and the direction of the interview guide.

Clinical outcomes of screening and diagnostic mammography in a limited resource healthcare system.

INTRODUCTION: Breast cancer is a significant public health concern in Jordan. It is the most common cancer among Jordanian women. Despite its high incidence and advanced stage at time of diagnosis, the uptake of breast cancer screening in Jordan is low. This study aims to compare clinical outcomes of both screening and diagnostic mammogram among women in Jordan. METHODS: A retrospective cohort of 1005 women who underwent mammography in breast imaging unit in a tertiary hospital in Jordan. It aimed to investigate outcomes of screening and diagnostic mammography. recall rates, clinical manifestations and cancer rates were investigated. RESULTS: A total of 1005 participants were involved and divided into screening group (n = 634) and diagnostic group (n = 371). Women in the diagnostic group were more likely to be younger, premenopausal, smokers with higher BMI. Among the screening group, 22.3% were labeled with abnormal mammogram, 26% recalled for ultrasound, 46 patients underwent tissue biopsy and a total of 12 patients had a diagnosis of breast carcinoma. Among the diagnostic group, the most commonly reported symptoms were a feeling of breast mass, mastalgia and nipple discharge. Abnormal mammogram was reported in 50.4% of women, a complementary ultrasound was performed for 205 patients. A diagnostic Tru-cut biopsy for 144 patients and diagnostic excisional biopsy for 17 patients were performed. A total of 131 had a diagnosis of carcinoma. CONCLUSION: With the high possibility of identifying a carcinoma in mammography among symptomatic women and low uptake of screening mammogram, efforts to increase awareness and improve access to screening services are crucial in reducing the burden of breast cancer in Jordan.

AsymMirai: Interpretable Mammography-based Deep Learning Model for 1-5-year Breast Cancer Risk Prediction.

Background Mirai, a state-of-the-art deep learning-based algorithm for predicting short-term breast cancer risk, outperforms standard clinical risk models. However, Mirai is a black box, risking overreliance on the algorithm and incorrect diagnoses. Purpose To identify whether bilateral dissimilarity underpins Mirai's reasoning process; create a simplified, intelligible model, AsymMirai, using bilateral dissimilarity; and determine if AsymMirai may approximate Mirai's performance in 1-5-year breast cancer risk prediction. Materials and Methods This retrospective study involved mammograms obtained from patients in the EMory BrEast imaging Dataset, known as EMBED, from January 2013 to December 2020. To approximate 1-5-year breast cancer risk predictions from Mirai, another deep learning-based model, AsymMirai, was built with an interpretable module: local bilateral dissimilarity (localized differences between left and right breast tissue). Pearson correlation coefficients were computed between the risk scores of Mirai and those of AsymMirai. Subgroup analysis was performed in patients for whom AsymMirai's year-over-year reasoning was consistent. AsymMirai and Mirai risk scores were compared using the area under the receiver operating characteristic curve (AUC), and 95% CIs were calculated using the DeLong method. Results Screening mammograms (n = 210 067) from 81 824 patients (mean age, 59.4 years ± 11.4 [SD]) were included in the study. Deep learning-extracted bilateral dissimilarity produced similar risk scores to those of Mirai (1-year risk prediction, r = 0.6832; 4-5-year prediction, r = 0.6988) and achieved similar performance as Mirai. For AsymMirai, the 1-year breast cancer risk AUC was 0.79 (95% CI: 0.73, 0.85) (Mirai, 0.84; 95% CI: 0.79, 0.89; P = .002), and the 5-year risk AUC was 0.66 (95% CI: 0.63, 0.69) (Mirai, 0.71; 95% CI: 0.68, 0.74; P < .001). In a subgroup of 183 patients for whom AsymMirai repeatedly highlighted the same tissue over time, AsymMirai achieved a 3-year AUC of 0.92 (95% CI: 0.86, 0.97). Conclusion Localized bilateral dissimilarity, an imaging marker for breast cancer risk, approximated the predictive power of Mirai and was a key to Mirai's reasoning. © RSNA, 2024 Supplemental material is available for this article See also the editorial by Freitas in this issue.

A multi-task fusion model based on a residual-Multi-layer perceptron network for mammographic breast cancer screening.

BACKGROUND AND OBJECTIVE: Deep learning approaches are being increasingly applied for medical computer-aided diagnosis (CAD). However, these methods generally target only specific image-processing tasks, such as lesion segmentation or benign state prediction. For the breast cancer screening task, single feature extraction models are generally used, which directly extract only those potential features from the input mammogram that are relevant to the target task. This can lead to the neglect of other important morphological features of the lesion as well as other auxiliary information from the internal breast tissue. To obtain more comprehensive and objective diagnostic results, in this study, we developed a multi-task fusion model that combines multiple specific tasks for CAD of mammograms. METHODS: We first trained a set of separate, task-specific models, including a density classification model, a mass segmentation model, and a lesion benignity-malignancy classification model, and then developed a multi-task fusion model that incorporates all of the mammographic features from these different tasks to yield comprehensive and refined prediction results for breast cancer diagnosis. RESULTS: The experimental results showed that our proposed multi-task fusion model outperformed other related state-of-the-art models in both breast cancer screening tasks in the publicly available datasets CBIS-DDSM and INbreast, achieving a competitive screening performance with area-under-the-curve scores of 0.92 and 0.95, respectively. CONCLUSIONS: Our model not only allows an overall assessment of lesion types in mammography but also provides intermediate results related to radiological features and potential cancer risk factors, indicating its potential to offer comprehensive workflow support to radiologists.

Positive Predictive Values of Abnormality Scores From a Commercial Artificial Intelligence-Based Computer-Aided Diagnosis for Mammography.

OBJECTIVE: Artificial intelligence-based computer-aided diagnosis (AI-CAD) is increasingly used in mammography. While the continuous scores of AI-CAD have been related to malignancy risk, the understanding of how to interpret and apply these scores remains limited. We investigated the positive predictive values (PPVs) of the abnormality scores generated by a deep learning-based commercial AI-CAD system and analyzed them in relation to clinical and radiological findings. MATERIALS AND METHODS: From March 2020 to May 2022, 656 breasts from 599 women (mean age 52.6 ± 11.5 years, including 0.6% [4/599] high-risk women) who underwent mammography and received positive AI-CAD results (Lunit Insight MMG, abnormality score ≥ 10) were retrospectively included in this study. Univariable and multivariable analyses were performed to evaluate the associations between the AI-CAD abnormality scores and clinical and radiological factors. The breasts were subdivided according to the abnormality scores into groups 1 (10-49), 2 (50-69), 3 (70-89), and 4 (90-100) using the optimal binning method. The PPVs were calculated for all breasts and subgroups. RESULTS: Diagnostic indications and positive imaging findings by radiologists were associated with higher abnormality scores in the multivariable regression analysis. The overall PPV of AI-CAD was 32.5% (213/656) for all breasts, including 213 breast cancers, 129 breasts with benign biopsy results, and 314 breasts with benign outcomes in the follow-up or diagnostic studies. In the screening mammography subgroup, the PPVs were 18.6% (58/312) overall and 5.1% (12/235), 29.0% (9/31), 57.9% (11/19), and 96.3% (26/27) for score groups 1, 2, 3, and 4, respectively. The PPVs were significantly higher in women with diagnostic indications (45.1% [155/344]), palpability (51.9% [149/287]), fatty breasts (61.2% [60/98]), and certain imaging findings (masses with or without calcifications and distortion). CONCLUSION: PPV increased with increasing AI-CAD abnormality scores. The PPVs of AI-CAD satisfied the acceptable PPV range according to Breast Imaging-Reporting and Data System for screening mammography and were higher for diagnostic mammography.

MOB-CBAM: A dual-channel attention-based deep learning generalizable model for breast cancer molecular subtypes prediction using mammograms.

BACKGROUND AND OBJECTIVE: Deep Learning models have emerged as a significant tool in generating efficient solutions for complex problems including cancer detection, as they can analyze large amounts of data with high efficiency and performance. Recent medical studies highlight the significance of molecular subtype detection in breast cancer, aiding the development of personalized treatment plans as different subtypes of cancer respond better to different therapies. METHODS: In this work, we propose a novel lightweight dual-channel attention-based deep learning model MOB-CBAM that utilizes the backbone of MobileNet-V3 architecture with a Convolutional Block Attention Module to make highly accurate and precise predictions about breast cancer. We used the CMMD mammogram dataset to evaluate the proposed model in our study. Nine distinct data subsets were created from the original dataset to perform coarse and fine-grained predictions, enabling it to identify masses, calcifications, benign, malignant tumors and molecular subtypes of cancer, including Luminal A, Luminal B, HER-2 Positive, and Triple Negative. The pipeline incorporates several image pre-processing techniques, including filtering, enhancement, and normalization, for enhancing the model's generalization ability. RESULTS: While identifying benign versus malignant tumors, i.e., coarse-grained classification, the MOB-CBAM model produced exceptional results with 99 % accuracy, precision, recall, and F1-score values of 0.99 and MCC of 0.98. In terms of fine-grained classification, the MOB-CBAM model has proven to be highly efficient in accurately identifying mass with (benign/malignant) and calcification with (benign/malignant) classification tasks with an impressive accuracy rate of 98 %. We have also cross-validated the efficiency of the proposed MOB-CBAM deep learning architecture on two datasets: MIAS and CBIS-DDSM. On the MIAS dataset, an accuracy of 97 % was reported for the task of classifying benign, malignant, and normal images, while on the CBIS-DDSM dataset, an accuracy of 98 % was achieved for the classification of mass with either benign or malignant, and calcification with benign and malignant tumors. CONCLUSION: This study presents lightweight MOB-CBAM, a novel deep learning framework, to address breast cancer diagnosis and subtype prediction. The model's innovative incorporation of the CBAM enhances precise predictions. The extensive evaluation of the CMMD dataset and cross-validation on other datasets affirm the model's efficacy.