Empowering breast cancer diagnosis and radiology practice: advances in artificial intelligence for contrast-enhanced mammography.

Artificial intelligence (AI) applications in breast imaging span a wide range of tasks including decision support, risk assessment, patient management, quality assessment, treatment response assessment and image enhancement. However, their integration into the clinical workflow has been slow due to the lack of a consensus on data quality, benchmarked robust implementation, and consensus-based guidelines to ensure standardization and generalization. Contrast-enhanced mammography (CEM) has improved sensitivity and specificity compared to current standards of breast cancer diagnostic imaging i.e., mammography (MG) and/or conventional ultrasound (US), with comparable accuracy to MRI (current diagnostic imaging benchmark), but at a much lower cost and higher throughput. This makes CEM an excellent tool for widespread breast lesion characterization for all women, including underserved and minority women. Underlining the critical need for early detection and accurate diagnosis of breast cancer, this review examines the limitations of conventional approaches and reveals how AI can help overcome them. The Methodical approaches, such as image processing, feature extraction, quantitative analysis, lesion classification, lesion segmentation, integration with clinical data, early detection, and screening support have been carefully analysed in recent studies addressing breast cancer detection and diagnosis. Recent guidelines described by Checklist for Artificial Intelligence in Medical Imaging (CLAIM) to establish a robust framework for rigorous evaluation and surveying has inspired the current review criteria.

Imaging of Benign and Malignant Breast Lesions Using Contrast-Enhanced Ultrasound: A Pictorial Essay.

ABSTRACT: Contrast-enhanced ultrasound is a promising noninvasive imaging technique for evaluating benign and malignant breast lesions, as contrast provides information about perfusion and microvasculature. Contrast-enhanced ultrasound is currently off-label use in the breast in the United States, but its clinical and investigational use in breast imaging is gaining popularity. It is important for radiologists to be familiar with the imaging appearances of benign and malignant breast masses using contrast-enhanced ultrasound. This pictorial essay illustrates enhancement patterns of various breast masses from our own experience. Pathologies include subtypes of invasive breast cancer, fibroadenomas, papillary lesions, fibrocystic change, and inflammatory processes. Contrast-enhanced ultrasound pitfalls and limitations are discussed.

Breast Density Legislation Impact on Breast Cancer Screening and Risk Assessment.

OBJECTIVE: To evaluate breast density notification legislation (BDNL) on breast imaging practice patterns, risk assessment, and supplemental screening. METHODS: A 20-question anonymous web-based survey was administered to practicing Society of Breast Imaging radiologists in the U.S. between February and April 2021 regarding breast cancer risk assessment, supplemental screening, and density measurements. Results were compared between facilities with and without BDNL using the two-sided Fisher's exact test. RESULTS: One hundred and ninety-seven radiologists from 41 U.S. states, with (187/197, 95%) or without (10/197, 5%) BDNL, responded. Fifty-seven percent (113/197) performed breast cancer risk assessment, and 93% (183/197) offered supplemental screening for women with dense breasts. Between facilities with or without BDNL, there was no significant difference in whether risk assessment was (P = 0.19) or was not performed (P = 0.20). There was no significant difference in supplemental screening types (P > 0.05) between BDNL and non-BDNL facilities. Thirty-five percent (69/197) of facilities offered no supplemental screening studies, and 25% (49/197) had no future plans to offer supplemental screening. A statistically significant greater proportion of non-BDNL facilities offered no supplemental screening (P < 0.03) and had no plans to offer supplemental screening compared to BDNL facilities (P < 0.02). CONCLUSION: Facilities in BDNL states often offer supplemental screening compared to facilities in non-BDNL states. Compared to BDNL facilities, a statistically significant proportion of non-BDNL facilities had no supplemental screening nor plans for implementation. Our data suggest that upcoming federal BDNL will impact how supplemental screening is addressed in currently non-BDNL states.

Multicenter Study of Whole Breast Stiffness Imaging by Ultrasound Tomography (SoftVue) for Characterization of Breast Tissues and Masses.

We evaluated whole breast stiffness imaging by SoftVue ultrasound tomography (UST), extracted from the bulk modulus, to volumetrically map differences in breast tissues and masses. A total 206 women with either palpable or mammographically/sonographically visible masses underwent UST scanning prior to biopsy as part of a prospective, HIPAA-compliant multicenter cohort study. The volumetric data sets comprised 298 masses (78 cancers, 105 fibroadenomas, 91 cysts and 24 other benign) in 239 breasts. All breast tissues were segmented into six categories, using sound speed to separate fat from fibroglandular tissues, and then subgrouped by stiffness into soft, intermediate and hard components. Ninety percent of women had mammographically dense breasts but only 11.2% of their total breast volume showed hard components while 69% of fibroglandular tissues were softer. All smaller masses (<1.5 cm) showed a greater percentage of hard components than their corresponding larger masses (p < 0.001). Cancers had significantly greater mean stiffness indices and lower mean homogeneity of stiffness than benign masses (p < 0.05). SoftVue stiffness imaging demonstrated small stiff masses, mainly due to cancers, amongst predominantly soft breast tissues. Quantitative stiffness mapping of the whole breast and underlying masses may have implications for screening of women with dense breasts, cancer risk evaluations, chemoprevention and treatment monitoring.

Accuracy of contrast-enhanced spectral mammography compared with MRI for invasive breast cancers: Prospective study in population of predominantly underrepresented minorities.

OBJECTIVES: This prospective study compares contrast-enhanced spectral mammography (CESM) with contrast-enhanced breast MRI in assessing the extent of newly diagnosed breast cancer in a multiethnic cohort. METHODS: This study includes 41 patients with invasive breast cancer detected by mammography or conventional ultrasound imaging from May 2017 to March 2020. CESM and MRI scans were performed prior to any treatment. Results are compared with each other and to histopathology. Detection of the malignant lesion was assessed by sensitivity, specificity, PPV, NPV. Consistency of malignant tumor size measurement was compared between modalities using Intraclass Correlation Coefficient (ICC). RESULTS: In a multiethnic cohort with over 65% Hispanic and African-American women, the sensitivity of detecting malignant lesions for CESM is 93.1% (77.23%, 99.15%) and MRI is 96.55% (82.24%, 99.91%). The PPV for CESM 96.43% (81.65%, 99.91%) is better compared to MRI 82.35% (65.47%, 93.24%). CESM is as effective as MRI in evaluating index cancers and multifocal/multicentric/contralateral disease. CESM has greater specificity and PPV since MRI tends to overcall benign lesions. There is a good agreement of tumor size between CESM to surgery and MRI to surgery with ICC of 0.85 (95% CI 0.69, 0.93) and 0.87 (95% CI 0.74, 0.94), respectively. There is good agreement of malignancy detection between CESM and MRI with Kappa of 0.74 (95% CI 0.52, 0.95). CONCLUSIONS: CESM is an effective imaging modality for evaluating the extent of disease in newly diagnosed invasive breast cancers and a good alternative to MRI in a multiethnic population.

Addressing ethnic disparities in imaging utilization and clinical outcomes for COVID-19.

PURPOSE: Racial and ethnic disparities have exacerbated during the COVID-19 pandemic as the healthcare system is overwhelmed. While Hispanics are disproportionately affected by COVID-19, little is known about ethnic disparities in the hospital settings. This study investigates imaging utilization and clinical outcomes between Hispanic and non-Hispanic COVID-19 patients in the Emergency Department (ED) and during hospitalization. METHODS: Through retrospective chart review, we included 331 symptomatic COVID-19 patients (mean age 53.2 years) at a metropolitan healthcare system from March to June 2020. Poisson regression was used to compare diagnostic imaging utilization and clinical outcomes between Hispanic and non-Hispanic patients. RESULTS: After adjusting for confounders, no statistically significant difference was found between Hispanic and non-Hispanic patients for the number of weekly chest X-rays. Results were categorized into four clinical outcomes: ED management (0.16 ± 0.05 vs. 0.14 ± 0.8, p:0.79); requiring inpatient management (1.31 ± 0.11 vs. 1.46 ± 0.16, p:0.43); ICU admission without invasive ventilation (1.4 ± 0.17 vs. 1.35 ± 0.26, p:0.86); and ICU admission and ventilator support (3.29 ± 0.22 vs. 3.59 ± 0.37, p:0.38). There were no statistically significant relative differences in adjusted prevalence rate between ethnic groups for all clinical outcomes (p > 0.05). There was a statistically significant longer adjusted length of stay (days) in non-Hispanics for two subcohorts: inpatient management (8.16 ± 0.31 vs. 9.72 ± 0.5, p < 0.01) and ICU admission without invasive ventilation (10.39 ± 0.57 vs. 13.45 ± 1.13, p < 0.01). CONCLUSIONS: For Hispanic and non-Hispanic COVID-19 patients in the ED or hospitalized, there were no statistically significant differences in imaging utilization and clinical outcomes.

The Fat-glandular Interface and Breast Tumor Locations: Appearances on Ultrasound Tomography Are Supported by Quantitative Peritumoral Analyses.

OBJECTIVE: To analyze the preferred tissue locations of common breast masses in relation to anatomic quadrants and the fat-glandular interface (FGI) using ultrasound tomography (UST). METHODS: Ultrasound tomography scanning was performed in 206 consecutive women with 298 mammographically and/or sonographically visible, benign and malignant breast masses following written informed consent to participate in an 8-site multicenter, Institutional Review Board-approved cohort study. Mass locations were categorized by their anatomic breast quadrant and the FGI, which was defined by UST as the high-contrast circumferential junction of fat and fibroglandular tissue on coronal sound speed imaging. Quantitative UST mass comparisons were done for each tumor and peritumoral region using mean sound speed and percentage of fibroglandular tissue. Chi-squared and analysis of variance tests were used to assess differences. RESULTS: Cancers were noted at the FGI in 95% (74/78) compared to 51% (98/194) of fibroadenomas and cysts combined (P < 0.001). No intra-quadrant differences between cancer and benign masses were noted for tumor location by anatomic quadrants (P = 0.66). Quantitative peritumoral sound speed properties showed that cancers were surrounded by lower mean sound speeds (1477 m/s) and percent fibroglandular tissue (47%), compared to fibroadenomas (1496 m/s; 65.3%) and cysts (1518 m/s; 84%) (P < 0.001; P < 0.001, respectively). CONCLUSION: Breast cancers form adjacent to fat and UST localized the vast majority to the FGI, while cysts were most often completely surrounded by dense tissue. These observations were supported by quantitative peritumoral analyses of sound speed values for fat and fibroglandular tissue.

Factors associated with MRI detection of occult lesions in newly diagnosed breast cancers.

BACKGROUND: The use of preoperative magnetic resonance imaging (MRI) for newly diagnosed breast cancer remains controversial. We examined factors associated with detection of occult multicentric, multifocal, and contralateral malignant lesions only seen by MRI. METHODS: We performed a retrospective analysis of consecutive patients undergoing preoperative MRI for breast cancer. Clinicopathologic data were assessed regarding the findings of multifocality, multicentricity, and the presence of contralateral lesions. We analyzed the association of factors with these findings on MRI. RESULTS: Of 857 patients undergoing MRI, 770 patients met inclusion criteria. Mean age was 54.7 years. Biopsy-proven detection rates by MRI for multifocal, multicentric, and contralateral cancers were 6.2% (48 of 770), 1.9% (15 of 770) and 3.1% (24 of 770), respectively. African American race and heterogeneously or extremely dense mammographic density were associated with multifocal cancers on MRI. Larger lesion size and mammographic density were associated with multicentric cancers. Invasive lobular carcinoma (ILC) and progesterone receptor (PR)-positivity were associated with contralateral cancers. CONCLUSIONS: African American race, heterogeneously or extremely dense mammographic density, ILC, and PR-positivity were associated with additional biopsy-proven cancers based on MRI. These factors should be considered when assessing the clinical utility of preoperative breast MRI.

Accuracy of Contrast-Enhanced Ultrasound Compared With Magnetic Resonance Imaging in Assessing the Tumor Response After Neoadjuvant Chemotherapy for Breast Cancer.

OBJECTIVES: This pilot study compared contrast enhanced ultrasound (US) with contrast-enhanced magnetic resonance imaging (MRI) in assessing the treatment response in patients with breast cancer receiving preoperative neoadjuvant chemotherapy (NAC). METHODS: This prospective Institutional Review Board-approved and Health Insurance Portability and Accountability Act-compliant study included 30 patients, from January 2014 to October 2015, with invasive breast cancer detected by mammography, conventional US imaging, or both and scheduled for NAC. Informed consent was obtained. Contrast-enhanced US (perflutren lipid microspheres, 10 µL/kg) and MRI (gadopentetate dimeglumine, 0.1 mmol/kg) scans were performed at baseline before starting NAC and after completing NAC before surgery. Results of the imaging techniques were compared with each other and with histopathologic findings obtained at surgery using the Spearman correlation. Tumor size and enhancement parameters were compared for 15 patients with contrast-enhanced US, MRI, and surgical pathologic findings. RESULTS: The median tumor size at baseline was 3.1 cm on both contrast-enhanced US and MRI scans. The Spearman correlation showed strong agreement in tumor size at baseline between contrast-enhanced US and MRI (r = 0.88; P < .001) but less agreement in tumor size after NAC (r = 0.66; P = .004). Trends suggested that contrast-enhanced US (r = 0.75; P < .001) had a better correlation than MRI (r = 0.42; P = .095) with tumor size at surgery. Contrast-enhanced US was as effective as MRI in predicting a complete pathologic response (4 patients; 75.0% accuracy for both) and a non-complete pathologic response (11 patients; 72.7% accuracy for both). CONCLUSIONS: Contrast enhanced US is a valuable imaging modality for assessing the treatment response in patients receiving NAC and had a comparable correlation as MRI with breast cancer size at surgery.

High-frequency ultrasound imaging for breast cancer biopsy guidance.

Image-guided core needle biopsy is the current gold standard for breast cancer diagnosis. Microcalcifications, an important radiographic finding on mammography suggestive of early breast cancer such as ductal carcinoma in situ, are usually biopsied under stereotactic guidance. This procedure, however, is uncomfortable for patients and requires the use of ionizing radiation. It would be preferable to biopsy microcalcifications under ultrasound guidance since it is a faster procedure, more comfortable for the patient, and requires no radiation. However, microcalcifications cannot reliably be detected with the current standard ultrasound imaging systems. This study is motivated by the clinical need for real-time high-resolution ultrasound imaging of microcalcifications, so that biopsies can be accurately performed under ultrasound guidance. We have investigated how high-frequency ultrasound imaging can enable visualization of microstructures in ex vivo breast tissue biopsy samples. We generated B-mode images of breast tissue and applied the Nakagami filtering technique to help refine image output so that microcalcifications could be better assessed during ultrasound-guided core biopsies. We describe the preliminary clinical results of high-frequency ultrasound imaging of ex vivo breast biopsy tissue with microcalcifications and without Nakagami filtering and the correlation of these images with the pathology examination by hematoxylin and eosin stain and whole slide digital scanning.

Radiologist's role in breast cancer staging: providing key information for clinicians.

Breast cancer is the second leading cause of cancer death in women, exceeded only by lung cancer, and the 5-year survival rate is largely dependent on disease stage. The American Joint Committee on Cancer (AJCC) staging system for breast cancer (7th edition) provides a tumor-node-metastasis (TNM) classification scheme for breast cancer that is important for determining prognosis and treatment. Ascertaining the correct stage of breast cancer can be challenging, and the importance of the radiologist's role has increased over the years. The radiologist should understand how breast cancer stage is assigned and should be familiar with the AJCC's TNM classification scheme. The authors review the AJCC's TNM staging system for breast cancer with emphasis on clinical and preoperative staging, the different imaging modalities used in staging, and the key information that should be conveyed to clinicians. Radiologic information that may alter stage, prognosis, or treatment includes tumor size; number of tumor lesions; total span of disease; regional nodal status (axillary levels I-III, internal mammary, supraclavicular); locoregional invasion (involvement of the pectoralis muscle, skin, nipple, or chest wall); and distant metastases to bone, lung, brain, and liver, among other anatomic structures.

The role of axillary ultrasound in the detection of metastases from primary breast cancers.

BACKGROUND: The value of diagnostic axillary ultrasound (AUS) in the preoperative evaluation of lymph nodes for breast cancer patients has yet to be completely clarified. METHODS: Results of AUS were reviewed for all patients with invasive cancers who were clinically node negative (cN0) and had subsequent axillary lymphadenectomy. Patients with positive ultrasound-guided node core biopsies bypassed sentinel lymph node biopsy (SLNB) and had axillary lymph node dissection, whereas those with sonographically normal nodes or benign/nondiagnostic biopsy results had SLNB. RESULTS: Of 128 cN0 patients with invasive cancer, 23 (18%) had abnormal axillary AUS. Of 18 core biopsies, 12 (67%) were malignant. SLNB was positive in 19 of 110 (17%) patients. ALND was performed in 32 (25%) patients. For determining axillary metastases, AUS sensitivity was 16 of 31 (52%), specificity was 90 of 97 (93%), the positive predictive value was 16 of 23 (69%), and the negative predictive value was 90 of 105 (86%). CONCLUSIONS: AUS examination was a valuable method for evaluating the axilla in newly diagnosed cN0 breast cancer patients.