RESUMEN

Background US is clinically established for breast imaging, but its diagnostic performance depends on operator experience. Computer-assisted (real-time) image analysis may help in overcoming this limitation. Purpose To develop precise real-time-capable US-based breast tumor categorization by combining classic radiomics and autoencoder-based features from automatically localized lesions. Materials and Methods A total of 1619 B-mode US images of breast tumors were retrospectively analyzed between April 2018 and January 2024. nnU-Net was trained for lesion segmentation. Features were extracted from tumor segments, bounding boxes, and whole images using either classic radiomics, autoencoder, or both. Feature selection was performed to generate radiomics signatures, which were used to train machine learning algorithms for tumor categorization. Models were evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity and were statistically compared with histopathologically or follow-up-confirmed diagnosis. Results The model was developed on 1191 (mean age, 61 years ± 14 [SD]) female patients and externally validated on 50 (mean age, 55 years ± 15]). The development data set was divided into two parts: testing and training lesion segmentation (419 and 179 examinations) and lesion categorization (503 and 90 examinations). nnU-Net demonstrated precision and reproducibility in lesion segmentation in test set of data set 1 (median Dice score [DS]: 0.90 [IQR, 0.84-0.93]; P = .01) and data set 2 (median DS: 0.89 [IQR, 0.80-0.92]; P = .001). The best model, trained with 23 mixed features from tumor bounding boxes, achieved an AUC of 0.90 (95% CI: 0.83, 0.97), sensitivity of 81% (46 of 57; 95% CI: 70, 91), and specificity of 87% (39 of 45; 95% CI: 77, 87). No evidence of difference was found between model and human readers (AUC = 0.90 [95% CI: 0.83, 0.97] vs 0.83 [95% CI: 0.76, 0.90]; P = .55 and 0.90 vs 0.82 [95% CI: 0.75, 0.90]; P = .45) in tumor classification or between model and histopathologically or follow-up-confirmed diagnosis (AUC = 0.90 [95% CI: 0.83, 0.97] vs 1.00 [95% CI: 1.00,1.00]; P = .10). Conclusion Precise real-time US-based breast tumor categorization was developed by mixing classic radiomics and autoencoder-based features from tumor bounding boxes. ClinicalTrials.gov identifier: NCT04976257 Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Bahl in this issue.

Asunto(s)

Asunto(s)

RESUMEN

The aim of this study was to investigate the measurement of the incident angle of the main blood vessel, and the benefits of its integral with ultrasound malignant features of breast nodules for the assessment of breast malignancy based on BI-RADS. The incident angles of main blood vessels of 185 breast nodules in 185 patients who underwent breast nodule surgical excision or biopsy were quantitatively measured using color Doppler ultrasound from October 2022 to October 2023 in a tertiary hospital, and related data were collected and analyzed. Based on histopathology as the gold standard, the breast nodules were classified into benign and malignant groups. The incident angle values of both groups were compared, Receiver Operating Characteristic (ROC) curves were plotted, and the optimal cutoff value for distinguishing between benign and malignant breast nodules was determined. The malignancy risk of the breast nodules was assessed using the incident angle of the breast main vessel, BI-RADS classification, and a combination of ultrasound malignant features with the incident angle. The areas under the ROC curves (AUC) for each method were calculated and compared. The average incident angle of the main vessel of the breast nodule for the benign and malignant breast nodule groups was (41.47 ± 14.27)° and (22.65 ± 11.09)°, respectively, with a statistically significant difference (t = 10.027, P < 0.001). There was a significant negative correlation between the breast nodule vessel incident angle and histopathological malignancy (r = - 0.593, P < 0.001). The ROC curve and Youden index suggested that the optimal cutoff value for distinguishing between benign and malignant breast nodules using the vascular incident angle was 26.9°, with a sensitivity of 76.34%, specificity of 84.78%, positive predictive value of 83.53%, negative predictive value of 78.00%, and AUC of 0.853. The diagnostic performance of BI-RADS classification based on ultrasound malignant features of the breast nodules alone in assessing the malignancy risk of breast nodules showed a sensitivity of 78.50%, specificity of 92.39%, positive predictive value of 91.25%, negative predictive value of 79.95%, and AUC of 0.905. The integral of ultrasound malignant features and vascular incident angle for BI-RADS based assessment for breast nodule malignancy risk demonstrated a sensitivity of 90.32%, specificity of 89.13%, positive predictive value of 89.36%, negative predictive value of 90.11%, and AUC of 0.940. The differences in negative predictive value and AUC between ultrasound malignant features BI-RADS classification alone and the combination of ultrasound malignant features BI-RADS classification with the incident angle of the main vessel of the breast nodule were all statistically significant (x2 = 3.243, P = 0.042; Z = 2.955, P = 0.003). Measuring the incident angle of the main blood vessel of breast nodules and combining this measurement with ultrasound malignant features for BI-RADS classification can enhance the effectiveness of malignancy risk assessment of breast nodules, increase the negative predictive value, and potentially reduce unnecessary biopsies.

Asunto(s)

RESUMEN

BACKGROUND: Fibroadenoma (FA) is documented as the most common benign breast disease typically presenting as a lump. A wide variety of other diseases including breast cancer can similarly present as lumps hence the need for further differentiation. Ultrasonography plays a vital role in the evaluation and treatment of breast lumps with histological analysis as the gold standard. OBJECTIVE: This study compared the physical and sonographic features of the breast in women with FA and women with breast lumps due to other diseases. MATERIALS AND METHODS: This is a single-centre comparative study. Clinical and sonographic breast evaluations of the recruited patients with lumps were done and reported using the American College of Radiology Breast Imaging Reporting and Data System score. The lumps were biopsied, and histological diagnosis was documented. Clinical and imaging features of the breasts of women with FA were then compared with those of women with lumps from other breast diseases, and collated data were analysed using SPSS Statistical version 23.0. RESULTS: Data from 118 subjects (59 in each group) were used for this study. There was a significant difference in the physical and sonographic appearance of FA concerning the patient's age, parity, change in lesion size, perilesional architecture, echogenicity, borders, capsule and background breast density. No FA was found in women with less dense breasts. CONCLUSION: The sonographic features of breasts showed some differences from the corresponding features of FA and other breast lesions. This has the potential to increase the efficiency of pre-operative diagnosis of FA and could be further applied in developing diagnostic criteria for FA in our environment.

Asunto(s)

RESUMEN

In our study, we evaluated the diagnostic performance of grayscale ultrasonography (USG) in risk stratification of mass-forming breast lesions. Our study included 90 cases, in which 88 were females and 2 cases were male with age ranging from 16 to 73 years. Out of 90 lesions, 51 (58%) lesions were benign and 39 (39%) lesions were malignant. High-resolution USG done in all 90 lesions revealed sensitivity (90.2%), specificity (74.36%), positive predictive value (PPV) (82.14%), negative predictive value (NPV) (85.29%), and accuracy (83.33%). Calculated weighted kappa value 0.665, indicating better level of agreement in predicting malignant lesions compared to gold standard. Our study revealed that USG is sensitive and specific test in detecting malignant lesions with high PPV and NPV; the calculated weighted kappa value was 0.665, indicating better level of agreement in predicting malignant lesions compared to gold standard.

RésuméDans notre étude, nous avons évalué les performances diagnostiques de l'échographie en niveaux de gris (USG) dans la stratification du risque de lésions mammaires formant une masse. Notre étude a inclus 90 cas, dont 88 femmes et 2 hommes âgés de 16 à 73 ans. Sur 90 lésions, 51 (58 %) étaient bénignes et 39 (39 %) étaient malignes. L'USG haute résolution réalisée sur les 90 lésions a révélé une sensibilité (90,2 %), une spécificité (74,36 %), une valeur prédictive positive (VPP) (82,14 %), une valeur prédictive négative (VPN) (85,29 %) et une précision (83,33 %). Valeur kappa pondérée calculée de 0,665, indiquant un meilleur niveau d'accord dans la prévision des lésions malignes par rapport à l'étalon-or. Notre étude a révélé que l'USG est un test sensible et spécifique pour détecter les lésions malignes avec une PPV et une VPN élevées; la valeur kappa pondérée calculée était de 0,665, ce qui indique un meilleur niveau de concordance dans la prévision des lésions malignes par rapport à l'étalon-or.

Asunto(s)

RESUMEN

PURPOSE: This study aimed to identify ultrasound and clinicopathological characteristics related to recurrence in HER2-positive (HER2+) breast cancer, and to develop nomograms for predicting recurrence. METHODS: In this dual-center study, we retrospectively enrolled 570 patients with HER2+ breast cancer. The ultrasound and clinicopathological characteristics of hormone receptor (HR)-/HER2+ patients and HR+/HER2+ patients were analyzed separately according to HR status. Eighty percent of the original samples from HR-/HER2+ and HR+/HER2+ patients were extracted by bootstrap sampling as the training cohorts, while the remaining 20% were used as the external validation cohorts. Informative characteristics were screened through univariate and multivariable Cox regression in the training cohorts and used to develop nomograms for predicting recurrence. The predictive accuracy was calculated using Harrell's C-index and calibration curves. RESULTS: Three informative characteristics (axillary nodal status, calcification, and Adler degree) were identified in HR-/HER2+ patients, and another three (histological grade, axillary nodal status, and echogenic halo) in HR+/HER2+ patients. Based on these, two separate nomograms were constructed to assess recurrence risk. In the training cohorts, the C-index was 0.740 (95% CI: 0.667-0.811) for HR-/HER2+ nomogram, and 0.749 (95% CI: 0.679-0.820) for HR+/HER2+ nomogram. In the validation cohorts, the C-index was 0.708 (95% CI: 0.540-0.877) for HR-/HER2+ group, and 0.705 (95% CI: 0.557-0.853) for HR+/HER2+ group. The calibration curves also indicated the excellent accuracy of the nomograms. CONCLUSIONS: Ultrasound performance of HER2+ breast cancers with different HR status was significantly different. Nomograms integrating ultrasound and clinicopathological characteristics exhibited favorable performance and have the potential to serve as a reliable method for predicting recurrence in heterogeneous breast cancer.

Asunto(s)

RESUMEN

Breast cancer is a leading cause of mortality among women globally, necessitating precise classification of breast ultrasound images for early diagnosis and treatment. Traditional methods using CNN architectures such as VGG, ResNet, and DenseNet, though somewhat effective, often struggle with class imbalances and subtle texture variations, leading to reduced accuracy for minority classes such as malignant tumors. To address these issues, we propose a methodology that leverages EfficientNet-B7, a scalable CNN architecture, combined with advanced data augmentation techniques to enhance minority class representation and improve model robustness. Our approach involves fine-tuning EfficientNet-B7 on the BUSI dataset, implementing RandomHorizontalFlip, RandomRotation, and ColorJitter to balance the dataset and improve model robustness. The training process includes early stopping to prevent overfitting and optimize performance metrics. Additionally, we integrate Explainable AI (XAI) techniques, such as Grad-CAM, to enhance the interpretability and transparency of the model's predictions, providing visual and quantitative insights into the features and regions of ultrasound images influencing classification outcomes. Our model achieves a classification accuracy of 99.14%, significantly outperforming existing CNN-based approaches in breast ultrasound image classification. The incorporation of XAI techniques enhances our understanding of the model's decision-making process, thereby increasing its reliability and facilitating clinical adoption. This comprehensive framework offers a robust and interpretable tool for the early detection and diagnosis of breast cancer, advancing the capabilities of automated diagnostic systems and supporting clinical decision-making processes.

Asunto(s)

RESUMEN

OBJECTIVES: To evaluate the added value of contrast-enhanced ultrasound (CEUS) on top of breast conventional imaging for predicting the upgrading of ductal carcinoma in situ (DCIS) to invasive cancer after surgery. METHODS: This retrospective study enrolled 140 biopsy-proven DCIS lesions in 138 patients and divided them into two groups based on postoperative histopathology: non-upgrade and upgrade groups. Conventional ultrasound (US), mammography (MMG), CEUS and clinicopathological (CL) features were reviewed and compared between the two groups. The predictive performance of different models (with and without CEUS features) for histologic upgrade were compared to calculate the added value of CEUS. RESULTS: Fifty-nine (42.1 %) lesions were histologically upgraded to invasive cancer after surgery. By logistic regression analyses, we found that high-grade DCIS at biopsy (P=0.004), ultrasonographic lesion size > 20 mm (P=0.007), mass-like lesion on US (P=0.030), the presence of suspicious calcification on MMG (P=0.014), the presence of perfusion defect (P=0.005) and the area under TIC>1021.34 ml (P<0.001) on CEUS were six independent factors predicting concomitant invasive components after surgery. The CL+US+MMG model made with the four predictors in the clinicopathologic, US and MMG categories yielded an area under the receiver operating curve (AUROC) value of 0.759 (95 % CI: 0.680-0.828) in predicting histological upgrade. The combination model built by adding the two CEUS predictors to the CL+US+MMG model showed higher predictive efficacy than the CL+US+MMG model (P=0.018), as the AUROC value was improved to 0.861 (95 % CI: 0.793-0.914). CONCLUSIONS: The addition of contrast-enhanced ultrasound to breast conventional imaging could improve the preoperative prediction of an upgrade to invasive cancer from CNB -proven DCIS lesions.

Asunto(s)

RESUMEN

Background It is unclear whether breast US screening outcomes for women with dense breasts vary with levels of breast cancer risk. Purpose To evaluate US screening outcomes for female patients with dense breasts and different estimated breast cancer risk levels. Materials and Methods This retrospective observational study used data from US screening examinations in female patients with heterogeneously or extremely dense breasts conducted from January 2014 to October 2020 at 24 radiology facilities within three Breast Cancer Surveillance Consortium (BCSC) registries. The primary outcomes were the cancer detection rate, false-positive biopsy recommendation rate, and positive predictive value of biopsies performed (PPV3). Risk classification of participants was performed using established BCSC risk prediction models of estimated 6-year advanced breast cancer risk and 5-year invasive breast cancer risk. Differences in high- versus low- or average-risk categories were assessed using a generalized linear model. Results In total, 34 791 US screening examinations from 26 489 female patients (mean age at screening, 53.9 years ± 9.0 [SD]) were included. The overall cancer detection rate per 1000 examinations was 2.0 (95% CI: 1.6, 2.4) and was higher in patients with high versus low or average risk of 6-year advanced breast cancer (5.5 [95% CI: 3.5, 8.6] vs 1.3 [95% CI: 1.0, 1.8], respectively; P = .003). The overall false-positive biopsy recommendation rate per 1000 examinations was 29.6 (95% CI: 22.6, 38.6) and was higher in patients with high versus low or average 6-year advanced breast cancer risk (37.0 [95% CI: 28.2, 48.4] vs 28.1 [95% CI: 20.9, 37.8], respectively; P = .04). The overall PPV3 was 6.9% (67 of 975; 95% CI: 5.3, 8.9) and was higher in patients with high versus low or average 6-year advanced cancer risk (15.0% [15 of 100; 95% CI: 9.9, 22.2] vs 4.9% [30 of 615; 95% CI: 3.3, 7.2]; P = .01). Similar patterns in outcomes were observed by 5-year invasive breast cancer risk. Conclusion The cancer detection rate and PPV3 of supplemental US screening increased with the estimated risk of advanced and invasive breast cancer. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Helbich and Kapetas in this issue.

Asunto(s)

RESUMEN

OBJECTIVE: The ultrasonographic characteristics of lymph node metastasis in breast cancer patients were retrospectively analyzed, and a predictive nomogram model was constructed to provide an imaging basis for better clinical evaluation. METHODS: B-mode ultrasound was used to retrospectively analyze the imaging characteristics of regional lymph nodes and tumors. Pathological examination confirmed the presence of lymph node metastasis in breast cancer patients. Univariable and multivariable logistic regression analyses were performed to analyze the risk factors for lymph node metastasis. LASSO regression analysis was performed to screen noninvasive indicators, and a nomogram prediction model was constructed for breast cancer patients with lymph node metastasis. RESULTS: A total of 187 breast cancer patients were enrolled, including 74 patients with lymph node metastasis in the positive group and 113 patients without lymph node metastasis in the negative group. Multivariate analysis revealed that pathological type (OR = 4.58, 95% CI: 1.44-14.6, p = 0.01), tumor diameter (OR = 1.37, 95% CI: 1.07-1.74, p = 0.012), spiculated margins (OR = 7.92, 95% CI: 3.03-20.67, p < 0.001), mixed echo of the breast tumor (OR = 37.09, 95% CI: 3.49-394.1, p = 0.003), and unclear lymphatic hilum structure (OR = 16.07, 95% CI: 2.41-107.02, p = 0.004) were independent risk factors for lymph node metastasis. A nomogram model was constructed for predicting breast cancer with lymph node metastasis, incorporating three significantly correlated indicators identified through LASSO regression analysis, namely, tumor spiculated margins, cortical thickness of lymph nodes, and unclear lymphatic hilum structure. The receiver operating characteristic (ROC) curve revealed that the area under the curve (AUC) was 0.717 (95% CI, 0.614-0.820) for the training set and 0.817 (95% CI, 0.738-0.890) for the validation set. The Hosmer-Lemeshow test results for the training set and the validation set were p = 0.9148 and p = 0.1648, respectively. The prediction nomogram has good diagnostic performance. CONCLUSIONS: B-mode ultrasound is helpful in the preoperative assessment of breast cancer patients with lymph node metastasis. The predictive nomogram model, which is based on logistic regression and LASSO regression analysis, is clinically safe, reliable, and highly practical.

Asunto(s)

RESUMEN

ABSTRACT: This study aims to explore the value of real-time strain elastography (RTE) and contrast-enhanced ultrasonography (CEUS) in the diagnosis of breast BI-RADS 4 lesions. It collected 85 cases (totaling 85 lesions) diagnosed with breast BI-RADS 4 through routine ultrasound from October 2020 to December 2022 in Huangshan City People's Hospital. All lesions underwent RTE and CEUS examination before surgery, and the ImageJ software was used to measure the periphery of lesion images in the enhancement peak mode and grayscale mode to calculate the contrast-enhanced ultrasound area ratio. The diagnostic capabilities of single-modal and multimodal ultrasound examination for the malignancy of breast BI-RADS 4 lesions were compared using the receiver operating characteristic curve; the Spearman correlation analysis was adopted to evaluate the correlation between multimodal ultrasound and CEUS area ratio. As a result, among the 85 lesions, 51 were benign, and 34 were malignant. The areas under the curve (AUCs) of routine ultrasound (US), US + RTE, US + CEUS, and US + RTE + CEUS were 0.816, 0.928, 0.953, and 0.967, respectively, with the combined method showing a higher AUC than the single application. The AUC of the CEUS area ratio diagnosing breast lesions was 0.888. There was a strong positive correlation (r = 0.819, P < 0.001) between the diagnostic performance of US + RTE + CEUS and the CEUS area ratio. In conclusion, based on routine ultrasound, the combination of RTE and CEUS can further improve the differential diagnosis of benign and malignant lesions in breast BI-RADS 4.

Asunto(s)

RESUMEN

ABSTRACT: The non-mass breast lesions on ultrasound (US) are a group of challenging pathology. We aimed to standardize these grayscale findings and investigate the effectiveness of superb microvascular imaging (SMI) and shear wave elastography (SWE). A total of 195 lesions were evaluated by B-mode US, SWE, and SMI in the same session. A "NON-MASS model" was built on grayscale US to group the lesions only as areas and those with associated features: microcalcifications, architectural distortion, ductal changes, and microcysts. The mean stiffness parameters Emean, Eratio, and mean vascular index (VI) were recorded following consecutive measurements. Besides, the microvascularity was graded based on Adler's classification (grades 0 to 3). Lesions were divided into 3 groups: benign, category B3, and malignant. One hundred twelve (57.4%) lesions were benign, 23 (11.8%) were B3, and 60 were (30.8%) in the malignant category. Thirty-eight (19.5%) lesions were observed only as an area, whereas associated features were present in 157 lesions (80.5%). Distortion was the only associated feature predicting malignancy among the grayscale findings (P < 0.001). There was a significant difference between malignant and nonmalignant (benign and B3) groups in terms of Adler's grade, Emean, Eratio, and VI values (P < 0.001). Sensitivity, specificity, and accuracy increased when advanced imaging parameters were added to grayscale findings (P < 0.001). In the presence of microcalcifications, architectural distortion, high elasticity, and hypervascularity in the "NON-MASS" imaging model, the suspicion of malignancy increases. The non-mass findings and advanced imaging techniques have the potential to find greater coverage in the following versions of BI-RADS atlas.

Asunto(s)

RESUMEN

PURPOSE: Acute nipple inversion can be unsettling for patients and is sometimes associated with an underlying breast malignancy. It also poses a diagnostic challenge with lack of consensus management guidelines. This study reviewed institutional experience with new nipple inversion, including malignant association, imaging utilization, and outcomes, in an effort to improve management. METHODS: A multisite institutional retrospective review was conducted of all breast imaging reports from 1/2010 to 6/2022 mentioning nipple inversion as an indication or finding. Patients with new nipple inversion, defined as arising since the time of last breast imaging exam or if reported as new by the patient/provider, were included for analysis. Retroareolar imaging findings, BI-RADS assessments/recommendations, pathology obtained from percutaneous or excisional biopsies, and follow-up imaging and clinical exams were collated. Cases of chronic or stable nipple inversion were excluded. Descriptive statistics were performed. RESULTS: A total of 414 patients had new nipple inversion, 387/414 (93.5 %) with benign or negative results at initial imaging and 27/414 (6.5 %) with malignant lesions. Diagnostic mammography/ultrasound detected 25/27 (92.6 %) cancers (sensitivity 92.6 %, specificity 75.5 %, PPV 20.8 %, NPV 99.3 %). Of 62 breast MRI exams performed in patients with negative mammogram/ultrasound, no cancers were detected in the retroareolar space with 2 incidental malignant lesions discovered distant from the nipple. CONCLUSION: Diagnostic mammography/ultrasound is reliable in workups of acute nipple inversion, with a high sensitivity and NPV for excluding malignancy. Breast MRI and surgical referral should be reserved for patients with suspicious associated symptoms or clinical findings.

Asunto(s)

RESUMEN

BACKGROUND: HER2 is a key biomarker for breast cancer treatment and prognosis. Traditional assessment methods like immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) are effective but costly and time-consuming. Our model incorporates these methods alongside photoacoustic imaging to enhance diagnostic accuracy and provide more comprehensive clinical insights. MATERIALS AND METHODS: A total of 301 breast tumors were included in this study, divided into HER2-positive (3+ or 2+ with gene amplification) and HER2-negative (below 3+ and 2+ without gene amplification) groups. Samples were split into training and validation sets in a 7:3 ratio. Statistical analyses involved t-tests, chi-square tests, and rank-sum tests. Predictive factors were identified using univariate and multivariate logistic regression, leading to the creation of three models: ModA (clinical factors only), ModB (clinical plus ultrasound factors), and ModC (clinical, ultrasound, and photoacoustic imaging-derived oxygen saturation (SO2)). RESULTS: The area under the curve (AUC) for ModA was 0.756 (95 % CI: 0.69-0.82), ModB increased to 0.866 (95 % CI: 0.82-0.91), and ModC showed the highest performance with an AUC of 0.877 (95 % CI: 0.83-0.92). These results indicate that the comprehensive model combining clinical, ultrasound, and photoacoustic imaging data (ModC) performed best in predicting HER2 expression. CONCLUSION: The findings suggest that integrating clinical, ultrasound, and photoacoustic imaging data significantly enhances the accuracy of predicting HER2 expression. For personalised breast cancer treatment, the integrated model could provide a comprehensive and reproducible decision support tool.

Asunto(s)

RESUMEN

OBJECTIVE: Identify the proportion of patients presenting for diagnostic breast imaging with clinically insignificant breast pain who are eligible for screening mammography and analyze the impact of routing these patients to screening on resource utilization, healthcare spending and cancer detection. METHODS: We retrospectively reviewed 100 consecutive women ≥40 years old without a history of breast cancer who underwent diagnostic mammogram and breast ultrasound for clinically insignificant breast pain from 1/2022 to 4/2022. Patients were screen-eligible if their last bilateral mammogram was over 12 months prior to presentation. Patients with only screening views during diagnostic mammography were assumed to have a negative/benign screening mammogram. Costs were calculated using the Centers for Medicare & Medicaid Services Physician Fee Schedule. RESULTS: 68 of 100 patients with breast pain were screen-eligible at time of diagnostic imaging. With a screen first approach, 47/68 would have had negative/benign screening mammograms, allowing for the availability of 47 diagnostic breast imaging appointments. The current workflow led to 100 diagnostic mammograms and ultrasounds, 29 follow-up ultrasounds, and 10 image-guided biopsies, with a total cost of $42,872.41. With a screen first approach, there would have been 68 screening mammograms, 53 diagnostic mammograms and ultrasounds, 10 follow-up ultrasounds, and 9 image-guided biopsies, with a total cost of $34,231.60. Two cancers were identified, both associated with suspicious mammographic findings. None would have been missed in a screen-first approach. DISCUSSION: Identifying screen-eligible patients with clinically insignificant breast pain and routing them to screening mammogram improves radiology resource allocation and decreases healthcare spending without missing any cancers.

Asunto(s)

RESUMEN

OBJECTIVE: To evaluate the clinical performance and financial costs of breast-specific gamma imaging (BSGI) as a biopsy-reducing problem-solving strategy in patients with inconclusive diagnostic imaging findings. METHODS: A retrospective analysis of all patients for whom BSGI was utilized for inconclusive imaging findings following complete diagnostic mammographic and sonographic evaluation between January 2013 and December 2018 was performed. Positive BSGI findings were correlated and biopsied with either US or stereotactic technique with confirmation by clip location and pathology. After a negative BSGI result, patients were followed for a minimum of 24 months or considered lost to follow-up and excluded (22 patients). Results of further imaging studies, biopsies, and pathology results were analyzed. Net savings of avoided biopsies were calculated based on average Medicare charges. RESULTS: Four hundred and forty female patients from 30 to 95 years (mean 55 years) of age were included in our study. BSGI demonstrated a negative predictive value (NPV) of 98.4% (314/319) and a positive predictive value for biopsy of 35.5% (43/121). The overall sensitivity was 89.6% (43/48), and the specificity was 80.1% (314/392). In total, 78 false positive but only 5 false negative BSGI findings were identified. Six hundred and twenty-one inconclusive imaging findings were analyzed with BSGI and a total of 309 biopsies were avoided. Estimated net financial savings from avoided biopsies were $646 897. CONCLUSION: In the management of patients with inconclusive imaging findings on mammography or ultrasonography, BSGI is a problem-solving imaging modality with high NPV that helps avoid costs of image-guided biopsies.

Asunto(s)

RESUMEN

BACKGROUND: Due to the superficial location, suspicious findings of the nipple-areolar complex (NAC) are not amenable to stereotactic or MRI-guided sampling and have historically necessitated surgical biopsy or skin-punch biopsy. There are limited reports of US-guided core biopsy of the nipple (US-CBN). OBJECTIVE: We report our nearly 3-year pilot experience with US-CBN at an academic breast imaging center. METHODS: An institutional review board-exempt and HIPAA-compliant retrospective review was performed. We assessed patient demographics, breast imaging characteristics, procedural data, pathology, and outcomes. RESULTS: Nine female patients aged 27 to 64 underwent US-CBN from January 2021 to October 2023. Initial imaging abnormalities included abnormal MRI enhancement, mammographic calcifications, and sonographic masses. After initial or second-look US, all imaging findings had sonographic correlates for biopsy specimens, the majority of which were sonographic masses (8/9). US-CBN was performed by 6 breast radiologists using a variety of devices. All biopsy specimen results were concordant with sonographic abnormalities, although 1 was considered discordant from the initial abnormality seen on MRI. There were no complications, and discomfort during the procedure was well-treated. Two patients (22%, 2/9) were diagnosed with malignancy. CONCLUSION: This pilot study demonstrated that US-CBN can be performed by a breast radiologist for definitive diagnosis of suspicious nipple abnormalities seen on breast imaging, avoiding surgery, and maintaining nipple integrity. In our population, 22% (2/9) of US-CBNs revealed malignancy.

Asunto(s)

RESUMEN

In this pilot study, we investigated the utility of handheld ultrasound-guided photoacoustic (US-PA) imaging probe for analyzing ex-vivo breast specimens obtained from female patients who underwent breast-conserving surgery (BCS). We aimed to assess the potential of US-PA in detecting biochemical markers such as collagen, lipids, and hemoglobin, and compare these findings with routine imaging modalities (mammography, ultrasound) and histopathology results, particularly across various breast densities. Twelve ex-vivo breast specimens were obtained from female patients with a mean age of 59.7 ± 9.5 years who underwent BCS. The tissues were illuminated using handheld US-PA probe between 700 and 1100 nm across all margins and analyzed for collagen, lipids, and hemoglobin distribution. The obtained results were compared with routine imaging and histopathological assessments. Our findings revealed that lipid intensity and distribution decreased with increasing breast density, while collagen exhibited an opposite trend. These observations were consistent with routine imaging and histopathological analyses. Moreover, collagen intensity significantly differed (P < 0.001) between cancerous and normal breast tissue, indicating its potential as an additional biomarker for risk stratification across various breast conditions. The study results suggest that a combined assessment of PA biochemical information, such as collagen and lipid content, superimposed on grey-scale ultrasound findings could aid in distinguishing between normal and malignant breast conditions, as well as assist in BCS margin assessment. This underscores the potential of US-PA imaging as a valuable tool for enhancing breast cancer diagnosis and management, offering complementary information to existing imaging modalities and histopathology.

Asunto(s)

RESUMEN

Today, doctors rely heavily on medical imaging to identify abnormalities. Proper classification of these abnormalities enables them to take informed actions, leading to early diagnosis and treatment. This paper introduces the "EfficientKNN" model, a novel hybrid deep learning approach that combines the advanced feature extraction capabilities of EfficientNetB3 with the simplicity and effectiveness of the k-Nearest Neighbors (k-NN) algorithm. Initially, EfficientNetB3, pre-trained on ImageNet, is repurposed to serve as a feature extractor. Subsequently, a GlobalAveragePooling2D layer is applied, followed by an optional Principal Component Analysis (PCA) to reduce dimensionality while preserving critical information. PCA is used selectively when deemed necessary. The extracted features are then classified using an optimized k-NN algorithm, fine-tuned through meticulous cross-validation.Our model underwent rigorous training using a curated dataset containing benign, malignant, and normal medical images. Data augmentation techniques, including rotations, shifts, flips, and zooms, were employed to help the model generalize and efficiently handle new, unseen data. To enhance the model's ability to identify the important features necessary for accurate predictions, the dataset was refined using segmentation and overlay techniques. The training utilized an ensemble of optimization algorithms-SGD, Adam, and RMSprop-with hyperparameters set at a learning rate of 0.00045, a batch size of 32, and up to 120 epochs, facilitated by early stopping to prevent overfitting.The results demonstrate that the EfficientKNN model outperforms traditional models such as VGG16, AlexNet, and VGG19 in terms of accuracy, precision, and F1-score. Additionally, the model showed better results compared to EfficientNetB3 alone. Achieving a 100 % accuracy rate on multiple tests, the EfficientKNN model has significant potential for real-world diagnostic applications. This study highlights the model's scalability, efficient use of cloud storage, and real-time prediction capabilities, all while minimizing computational demands.By integrating the strengths of EfficientNetB3's deep learning architecture with the interpretability of k-NN, EfficientKNN presents a significant advancement in medical image classification, promising improved diagnostic accuracy and clinical applicability.

Asunto(s)

RESUMEN

PURPOSE: To create a simple model using standard BI-RADS® descriptors from pre-treatment B-mode ultrasound (US) combined with clinicopathological tumor features, and to assess the potential of the model to predict the presence of residual tumor after neoadjuvant chemotherapy (NAC) in breast cancer (BC) patients. METHOD: 245 female BC patients receiving NAC between January 2017 and December 2019 were included in this retrospective study. Two breast imaging fellows independently evaluated representative B-mode tumor images from baseline US. Additional clinicopathological tumor features were retrieved. The dataset was split into 170 training and 83 validation cases. Logistic regression was used in the training set to identify independent predictors of residual disease post NAC and to create a model, whose performance was evaluated by ROC curve analysis in the validation set. The reference standard was postoperative histology to determine the absence (pathological complete response, pCR) or presence (non-pCR) of residual invasive tumor in the breast or axillary lymph nodes. RESULTS: 100 patients (40.8%) achieved pCR. Logistic regression demonstrated that tumor size, microlobulated margin, spiculated margin, the presence of calcifications, the presence of edema, HER2-positive molecular subtype, and triple-negative molecular subtype were independent predictors of residual disease. A model using these parameters demonstrated an area under the ROC curve of 0.873 in the training and 0.720 in the validation set for the prediction of residual tumor post NAC. CONCLUSIONS: A simple model combining standard BI-RADS® descriptors from pre-treatment B-mode breast US with clinicopathological tumor features predicts the presence of residual disease after NAC.

Asunto(s)