Malignant Breast Lesions Research Articles

Predicting malignancy in breast lesions: enhancing accuracy with fine-tuned convolutional neural network models.

This study aims to explore the accuracy of Convolutional Neural Network (CNN) models in predicting malignancy in Dynamic Contrast-Enhanced Breast Magnetic Resonance Imaging (DCE-BMRI). A total of 273 benign lesions (benign group) and 274 malignant lesions (malignant group) were collected and randomly divided into a training set (246 benign and 245 malignant lesions) and a testing set (28 benign and 28 malignant lesions) in a 9:1 ratio. An additional 53 lesions from 53 patients were designated as the validation set. Five models-VGG16, VGG19, DenseNet201, ResNet50, and MobileNetV2-were evaluated. Model performance was assessed using accuracy (Ac) in the training and testing sets, and precision (Pr), recall (Rc), F1 score (F1), and area under the receiver operating characteristic curve (AUC) in the validation set. The accuracy of VGG19 on the test set (0.96) is higher than that of VGG16 (0.91), DenseNet201 (0.91), ResNet50 (0.67), and MobileNetV2 (0.88). For the validation set, VGG19 achieved higher performance metrics (Pr 0.75, Rc 0.76, F1 0.73, AUC 0.76) compared to the other models, specifically VGG16 (Pr 0.73, Rc 0.75, F1 0.70, AUC 0.73), DenseNet201 (Pr 0.71, Rc 0.74, F1 0.69, AUC 0.71), ResNet50 (Pr 0.65, Rc 0.68, F1 0.60, AUC 0.65), and MobileNetV2 (Pr 0.73, Rc 0.75, F1 0.71, AUC 0.73). S4 model achieved higher performance metrics (Pr 0.89, Rc 0.88, F1 0.87, AUC 0.89) compared to the other four fine-tuned models, specifically S1 (Pr 0.75, Rc 0.76, F1 0.74, AUC 0.75), S2 (Pr 0.77, Rc 0.79, F1 0.75, AUC 0.77), S3 (Pr 0.76, Rc 0.76, F1 0.73, AUC 0.75), and S5 (Pr 0.77, Rc 0.79, F1 0.75, AUC 0.77). Additionally, S4 model showed the lowest loss value in the testing set. Notably, the AUC of S4 for BI-RADS 3 was 0.90 and for BI-RADS 4 was 0.86, both significantly higher than the 0.65 AUC for BI-RADS 5. The S4 model we propose has demonstrated superior performance in predicting the likelihood of malignancy in DCE-BMRI, making it a promising candidate for clinical application in patients with breast diseases. However, further validation is essential, highlighting the need for additional data to confirm its efficacy.

Benign and malignant breast lesions in children and adolescents - diagnostic and therapeutic approach.

Benign and malignant breast lesions in children and adolescents are rare compared to adults. Most tumors are benign. Malignant breast lesions are extremely rare. Fibroadenomas are the most common, accounting for 95% of all lesions. Diagnosis is based on history and physical examination of the breast and armpit. Imaging studies include ultrasound, mammography, and magnetic resonance imaging. Ultrasound is the most commonly used imaging test. Other tests are used in cases of diagnostic doubt. Core needle biopsy should be considered for appropriate diagnostic management. Excisional biopsy should be considered for complex clinical conditions and imaging studies. Except in doubtful situations in children and adolescent girls, a conservative approach and observation of the lesions along with periodic ultrasound examination initially every 6-12 months is advisable. Management of malignant breast lesions in children typically involves a multidisciplinary team consisting of pediatric oncologists, surgeons, radiation oncologists, pathologists, and other specialists and depends on the clinical condition of the patient. An important aspect is the experience of the clinician and radiologist in the treatment of breast lesions, as well as increasing patient and family awareness of possible breast lesions and self-examination. This review aims to provide a scoping overview of the available literature on benign and malignant lesions of the breast in pediatric and adolescent populations to assist physicians and surgeons in making decisions regarding the appropriate diagnosis and management of pediatric breast disease.

Study on the classification of benign and malignant breast lesions using a multi-sequence breast MRI fusion radiomics and deep learning model

PurposeTo develop a multi-modal model combining multi-sequence breast MRI fusion radiomics and deep learning for the classification of benign and malignant breast lesions, to assist clinicians in better selecting treatment plans. MethodsA total of 314 patients who underwent breast MRI examinations were included. They were randomly divided into training, validation, and test sets in a ratio of 7:1:2. Subsequently, features of T1-weighted images (T1WI), T2-weighted images (T2WI), and dynamic contrast-enhanced MRI (DCE-MRI) were extracted using the convolutional neural network ResNet50 for fusion, and then combined with radiomic features from the three sequences. The following models were established: T1 model, T2 model, DCE model, DCE_T1_T2 model, and DCE_T1_T2_rad model. The performance of the models was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. The differences between the DCE_T1_T2_rad model and the other four models were compared using the Delong test, with a P-value < 0.05 considered statistically significant. ResultsThe five models established in this study performed well, with AUC values of 0.53 for the T1 model, 0.62 for the T2 model, 0.79 for the DCE model, 0.94 for the DCE_T1_T2 model, and 0.98 for the DCE_T1_T2_rad model. The DCE_T1_T2_rad model showed statistically significant differences (P < 0.05) compared to the other four models. ConclusionThe use of a multi-modal model combining multi-sequence breast MRI fusion radiomics and deep learning can effectively improve the diagnostic performance of breast lesion classification.

The Utility of Contrast-Enhanced Mammography in the Evaluation of Bloody Nipple Discharge-A Multicenter Study in the Asian Population.

To assess the efficacy of contrast-enhanced mammography (CEM) in differentiating benign from malignant breast lesions in Asian patients with bloody nipple discharge (BND). This retrospective study included 58 women with BND (mean age: 51.7 years) who underwent standardized CEM at institutions in Taiwan and Singapore. Lesion characteristics (size, enhancement, conspicuity, shape, margins) were evaluated on CEM by blinded radiologists. Non-enhanced mammography (MMG) and ultrasound (US) within a defined timeframe were compared for diagnostic accuracy. Benign or malignant status was confirmed by biopsy or 2-year imaging follow-up. Malignancy was found in 29 of 58 lesions (50.0%), with ductal carcinoma in situ (DCIS) being the most common. CEM demonstrated a 100% negative predictive value (NPV) for non-enhancing lesions. Significant predictors of malignancy on multivariate analysis include enhancing lesions of size ≥ 1.5 cm (p-value 0.025) and suspicious morphological features (irregular/spiculated margins, irregular shape, segmental/linear NME distribution) (p-value < 0.001). CEM outperformed MMG (sensitivity: 58.6%) and US (sensitivity: 79.3%), achieving a sensitivity of 100% and the highest diagnostic accuracy at 81.3%. Additionally, a CEM size cut-off of 1.5 cm yielded a sensitivity of 73.5% and a specificity of 84.3%. CEM effectively differentiates benign from malignant lesions in patients with BND, improving diagnostic accuracy and potentially reducing unnecessary interventions.

Intraoperative evaluation of tumor margins using a TROP2 near-infrared imaging probe to enable human breast-conserving surgery.

Intraoperative surgical margin assessment remains a challenge during breast-conserving surgery. Here, we report a combined strategy of immuno-positron emission tomography (PET) for preoperative detection of breast cancer and guided assessment of margins in breast-conserving surgery through second near-infrared (NIR-II) fluorescence imaging of trophoblastic cell surface antigen 2 (TROP2). We demonstrated that the intensity of PET signals in the tumors was nearly five times higher than in normal breast tissue with a zirconium-89 tracer conjugated to sacituzumab govitecan (SG) in a mouse spontaneous breast cancer model, enabling the identification of tumors. We further generated a NIR-II probe of indocyanine green conjugated to SG (ICG-SG) and developed a rapid incubation imaging method for intraoperative margin assessment in a relevant time window for the operation workflow. The ICG-SG NIR-II fluorescence image guidance was first verified to remove tumors completely and accurately in mouse breast cancer models. Moreover, the rapid incubation imaging method was applied to distinguish benign and malignant breast lesions in samples from 26 patients with breast cancer. Therefore, we have developed both nuclide and optical probes targeting TROP2 for rapid and precise identification of tumor margins during breast-conserving surgery in humans.

Multimodal apparent diffusion MRI model in noninvasive evaluation of breast cancer and Ki-67 expression

BackgroundTo assess the capability of multimodal apparent diffusion (MAD) weighted magnetic resonance imaging (MRI) to distinguish between malignant and benign breast lesions, and to predict Ki-67 expression level in breast cancer.MethodsThis retrospective study was conducted with 93 patients who had postoperative pathology-confirmed breast cancer or benign breast lesions. MAD images were acquired using a 3.0 T MRI scanner with 16 b values. The MAD parameters, as flow (fF, DF), unimpeded (fluid) (fUI), hindered (fH, DH, and αH), and restricted (fR, DR), were calculated. The differences of the parameters were compared by Mann–Whitney U test between the benign/malignant lesions and high/low Ki-67 expression level. The diagnostic performance was assessed by the area under the receiver operating characteristic curve (AUC).ResultsThe fR in the malignant lesions was significantly higher than in the benign lesions (P = 0.001), whereas the fUI and DH were found to be significantly lower (P = 0.007 and P < 0.001, respectively). Compared with individual parameter in differentiating malignant from benign breast lesions, the combination parameters of MAD (fR, DH, and fUI) provided the highest AUC (0.851). Of the 73 malignant lesions, 42 (57.5%) were assessed as Ki-67 low expression and 31 (42.5%) were Ki-67 high expression. The Ki-67 high status showed lower DH, higher DF and higher αH (P < 0.05). The combination parameters of DH, DF, and αH provided the highest AUC (0.691) for evaluating Ki-67 expression level.ConclusionsMAD weighted MRI is a useful method for the breast lesions diagnostics and the preoperative prediction of Ki-67 expression level.

Utility of diffusion-weighted imaging in differentiating benign and malignant breast lesions.

Breast cancer presents a significant global health burden. An accurate differentiation between benign and malignant lesions is imperative for timely intervention. While dynamic contrast enhanced MRI (DCE-MRI) is highly sensitive, its specificity is limited. This has led to the exploration of diffusion-weighted imaging (DWI) in distinguishing between benign and malignant breast lesions. The study aimed to explore the diagnostic utility of DWI in distinguishing between benign and malignant breast lesions. Assessment of 38 breast lesions using DWI with a b value of 800 s/mm2, performed with 3 Tesla MRI. The diagnostic performance of two different region of Interest (ROI) placement approaches was compared to obtain a feasible cut-off value of apparent diffusion coefficient (ADC) to differentiate between malignant and benign lesions. The histopathological reports were used as the gold standard. ADC values of malignant lesions were significantly lower than those of benign lesions (0.84 × 10-3 mm2/s vs. 1.54 × 10-3 mm2/s). The average ADC measured using a small-sized 2D ROI including the darkest part in the ADC map, performed better than the large 2D ROI covering the entire lesion. Using a cut-off value of 0.98 × 10-3 mm2/s, ADC obtained high sensitivity (90%) and specificity (88.9%) in distinguishing between benign and malignant breast lesions. Utilising quantitative analysis of DWI with ADC value measurement, reliably distinguished between benign and malignant breast lesions in this cohort, especially when employing a higher b value of 800 s/mm2.

Diffuse correlation tomography: a technique to characterize tissue blood flow abnormalities in benign and malignant breast lesions

Accurate assessment and quantification of neoangiogenesis associated with breast cancer could be potentially used to improve the sensitivity and specificity of non-invasive diagnosis, as well as predict outcomes and monitor treatment effects. In this study, we adapted an emerging technology, namely diffuse correlation tomography (DCT), to image microvascular blood flow in breast tissues and evaluate the potential for discriminating between benign and malignant lesions. A custom-made DCT system was designed for breast blood flow imaging, with both the source-detector array and reconstruction algorithm optimized to ensure precise imaging of breast blood flow. The global features and local features of three-dimensional blood flow images were extracted from the relative blood flow index (rBFI), which was obtained from most of the breasts targeted to the lesion. A total of 37 women with 19 benign and 18 malignant lesions were included in the study. Significant differences between malignant and benign groups were found in 12 image features. Moreover, when selecting the lesion mean relative blood flow index (MrBFI) as a single indicator, the malignant and benign tumors were discriminated with an accuracy of 89.2%. The blood flow features were found to successfully identify malignant and benign tumors, suggesting that DCT, as an alternate functional imaging modality, has the potential to be translated into clinical practice for diagnosis and assessment of breast cancers. There is potential to reduce the need for biopsy of benign lesions by improving the specificity of diagnostic imaging, as well as monitoring response to breast cancer treatment.

What Is the Added Value of DWI Compared With Structured Assessment of BI-RADS Criteria by the Kaiser Score? A Systematic Review and Meta-analysis

Objective This systematic review and meta-analysis investigated the added value of DWI compared with the structured assessment of BI-RADS criteria using the Kaiser score. Materials and Methods Articles published in English until May 2024 were included. Two independent reviewers extracted data on the characteristics of studies evaluating the added value of DWI to distinguish benign from malignant breast lesions compared with structured assessment of the BI-RADS criteria. Using bivariate random-effects models, the sensitivity and specificity were calculated. I 2 statistics, Deek's funnel plot asymmetry test for publication bias, and meta-regression were applied for the data analysis. Results Five studies comprising 1005 malignant and 846 benign lesions were eligible for data synthesis. The pooled sensitivity and specificity estimates of structured BI-RADS assessment were 95.7% (95% confidence interval [CI], 92.6%–97.5%) and 68.7% (95% CI, 60.9%–75.6%), respectively. Adding DWI to the structured BI-RADS assessment achieved a pooled sensitivity of 94.4% (95% CI, 90.5%–96.7%) and a pooled specificity of 74.9% (95% CI, 68.8%–80.2%). Adding DWI to the structured BI-RADS assessment significantly changed neither the sensitivity (P = 0.52) nor the specificity (P = 0.20). Conclusions This systematic review and meta-analysis revealed only a limited, statistically nonsignificant added value of DWI compared with the structured assessment of BI-RADS criteria using the Kaiser score.

Machine learning-based discrimination of benign and malignant breast lesions on US: The contribution of shear-wave elastography

PurposeTo build and validate a combined radiomics and machine learning (ML) approach using B-mode US and SWE images to differentiate benign from malignant solid breast lesions (BLs) and compare its performance with that of an expert radiologist. MethodsPatients with at least one BI-RADS 2–6 BL who performed breast US integrated with SWE were retrospectively included. B-mode US and SWE images were manually segmented to extract radiomics features. A multi-step feature selection process was performed and a predictive model built using the Logistic Regression algorithm. The diagnostic accuracy was evaluated with the AUC and Matthews Correlation Coefficient (MCC) metrics. The performance of the ML classifier was compared to that of an expert radiologist. Results427 Bls were included and divided into a training (286 BLs, of which 127 benign and 159 malignant) and a test set (141 BLs, of which 59 benign and 82 malignant). Of 1098 features extracted from B-mode US and SWE images, 13 were finally selected. The ML classifier showed an AUC of 0.768 and 0.746, and an MCC of 0.403 and 0.423 in the training and test sets, respectively. The performance was higher than that of the expert radiologist assessing only B-mode US images, but significantly lower when SWE images were also provided. ConclusionA ML approach based on B-mode US and SWE images may represent a potential tool in the characterization of BLs. SWE still gives its most relevant contribution in the clinical setting rather than included in a radiomics pipeline.

The value and significance of nucleolar organizer region proteins as markers of malignancy in breast cancer patients.

To assess argyrophilic nucleolar organizer regions (AgNORs) in 60 patients with primary breast carcinoma and evaluated their association with clinical prognostic parameters of breast cancer. Argyrophilic nucleolar organizer regions were stained in paraffin sections of the tissues using Ploton's silver method. For each sample, the number of AgNORs within the nuclei of 100 tumor cells was counted. The average number of AgNORs per nucleus was calculated, and the results were expressed as mean. The number of AgNORs was significantly higher in breast invasive ductal carcinoma (6.6) compared to benign breast tumors (fewer than 2.0). However, differences in AgNOR counts across different age groups were not statistically significant. This study suggests that AgNOR counts could be used as a potential procedure for estimating proliferation characteristics in histopathological sections of benign and malignant breast lesions. Argyrophilic nucleolar organizer region counts may also be valuable for identifying high-risk patients and indicating tumor aggressiveness. A larger study with an increased sample size that incorporates both AgNOR numbers and Ki67 scores for assessing cell kinetics is needed to confirm our findings.

Accuracy of distinguishing benign, high-risk lesions and malignancies with inductive machine learning models in BIRADS 4 and BIRADS 5 lesions on breast MR examinations

Rationale and ObjectivesThe aim of this study is to explore the utility of Inductive Decision Tree models (IDTs) in distinguishing between benign, malignant, and high-risk (B3) breast lesions. Materials and MethodsWe analyzed 124 histologically confirmed lesions in 114 patients who underwent breast MR with BI-RADS 4 or 5 assessment. The dataset comprised 10 imaging parameters and one clinical observation. Using the IDTs method (algorithm C5.0) combined with the data balancing method SMOTE (Synthetic Minority Oversampling Technique) and a corresponding new method called LCC (Leveling of Cases per Class), we developed corresponding 3-class classification models (Benign, B3, or Malignant). The training set used for classification model development consists of 112 cases with 11 variables, and the model’s performance was assessed using 10-fold Cross-Validation and Leave One Out methods (utilizing the training set) and the Use Test Set method (testing on an unknown (for the models) dataset of 12 cases with 12 variables). ResultsThis preliminary study demonstrates the potential for IDTs to accurately distinguish between benign, B3 and Malignant lesions based on extracted data from breast MRI exams with a high classification accuracy (88.70%), mean sensitivity of 97.18% and specificity of 98.59% achieved by the optimal classification model, derived from the combination of the IDTs method and the LCC data balancing method. ConclusionThe present study represents a pioneering approach for three-category classification of breast lesions in MRI images using Artificial Intelligence (AI) which can be utilized as second-opinion tool in daily clinical practice by radiologists.

Lesion conspicuity and contrast kinetics as predictors to differentiate benign and malignant breast lesions in contrast-enhanced mammogram

BackgroundContrast-enhanced mammography (CEM) is a recently developed, cost-effective imaging technique that offers both anatomical and functional breast imaging. Lesion conspicuity, a newly introduced lexicon in the ACR BIRADS supplementary atlas on CEM (2022), lacks sufficient data to correlate with malignancy likelihood. The feasibility of assessing contrast kinetics with CEM remains uncertain, and there is a scarcity of available data. Our research aims to address these gaps.ResultsTwo radiologists, blinded to pathological reports, independently evaluated 504 CEM enhanced breast lesions with histopathology reports, out of which 176 were benign and 328 were malignant. Subjective qualitative assessment of lesion conspicuity and contrast kinetics was done for each enhancing lesion. The lesion conspicuity was classified as low, moderate, or high. The kinetic behavior of each lesion was categorized into either persistent, plateau, or washout. The distribution of lesion conspicuity among benign and malignant lesions, respectively, was as follows: for low conspicuity, 74.4% versus 25.6%; for moderate conspicuity, 30.6% versus 69.4%; and for high conspicuity, 8.4% versus 91.6%. Regarding contrast kinetics and their distribution between benign and malignant lesions, persistent kinetics was detected in 95.6% compared to 4.4%, plateau kinetics in 43.4% versus 56.6%, and washout kinetics in 3.5% versus 96.5%. Statistically significant differences in distribution between benign and malignant lesions were observed for both lexicons (P < 0.001). The inter-observer agreement for lesion conspicuity (kappa = 0.97) and contrast kinetics (kappa = 0.92) was deemed excellent.ConclusionThe addition of lesion conspicuity and contrast kinetics as lexicons in CEM could enhance its diagnostic accuracy.

Diagnostic Accuracy of Mammography and Ultrasonography Screening for Breast Cancer in Pregnant and Lactating Women

Breast cancer was of significant health concern affecting women worldwide. Objective: To assess diagnostic accuracy of mammography and ultrasonography in differentiating malignant and benign breast lesions in pregnant and lactating women. Methods: A cross sectional retrospective study was conducted at Radiology department of Shahida Islam Medical Complex, Lodhran from May 2023 to April 2024. A sample size of 242 females was calculated. Electronic medical records were reviewed for radiological examination including screening ultrasound and mammography. All those pregnant and lactating females on which ultrasound and mammograms were performed were included. SPSS version 23.0 was used for data analysis. Diagnostic accuracy of both ultrasound and mammography were calculated in terms of malignant or benign and sensitivity and specificity. Mann-Whitney U test was applied between mammography, ultrasonography and BI-RADS categories. Results: Of 242 females, 110 underwent mammography and 132 underwent ultrasound. Negative mammography was observed in 71 females in which at biopsy, 24 were benign. Negative ultrasound with only benign lesion was seen in 68 females and 10 of which were confirmed at biopsy. Specificity of ultrasonography was 100 %, sensitivity 85.7 %, positive predictive value of 100 % while negative predictive value of 25 %. Specificity of mammography was 100 %, sensitivity 92.3 %, positive predictive value of 100 % while negative predictive value of 42.8 % (p&lt;0.001). Conclusions: Although both ultrasound and mammography were found to be specific, use of mammography was considered better in terms of sensitivity and diagnostic accuracy.

Enhancing Breast Cancer Detection through Advanced AI-Driven Ultrasound Technology: A Comprehensive Evaluation of Vis-BUS.

This study aims to enhance breast cancer detection accuracy through an AI-driven ultrasound tool, Vis-BUS, developed by Barreleye Inc., Seoul, South Korea. Vis-BUS incorporates Lesion Detection AI (LD-AI) and Lesion Analysis AI (LA-AI), along with a Cancer Probability Score (CPS), to differentiate between benign and malignant breast lesions. A retrospective analysis was conducted on 258 breast ultrasound examinations to evaluate Vis-BUS's performance. The primary methods included the application of LD-AI and LA-AI to b-mode ultrasound images and the generation of CPS for each lesion. Diagnostic accuracy was assessed using metrics such as the Area Under the Receiver Operating Characteristic curve (AUROC) and the Area Under the Precision-Recall curve (AUPRC). The study found that Vis-BUS achieved high diagnostic accuracy, with an AUROC of 0.964 and an AUPRC of 0.967, indicating its effectiveness in distinguishing between benign and malignant lesions. Logistic regression analysis identified that 'Fatty' lesion density had an extremely high odds ratio (OR) of 27.7781, suggesting potential convergence issues. The 'Unknown' density category had an OR of 0.3185, indicating a lower likelihood of correct classification. Medium and large lesion sizes were associated with lower likelihoods of correct classification, with ORs of 0.7891 and 0.8014, respectively. The presence of microcalcifications showed an OR of 1.360. Among Breast Imaging-Reporting and Data System categories, category C5 had a significantly higher OR of 10.173, reflecting a higher likelihood of correct classification. Vis-BUS significantly improves diagnostic precision and supports clinical decision-making in breast cancer screening. However, further refinement is needed in areas like lesion density characterization and calcification detection to optimize its performance.

EVALUATION OF FOCAL BREAST LESIONS USING ULTRASOUND ELASTOGRAPHY

Breast lesions pose a significant diagnostic challenge, necessitating advanced imaging techniques for accurate characterization. Ultrasound elastography, including strain and shear wave methods, offers promising diagnostic capabilities for distinguishing between benign and malignant breast lesions. Objective: This study aimed to evaluate the effectiveness of ultrasound elastography in characterizing focal breast lesions and to compare the diagnostic performance of strain and shear wave elastography techniques. Methods: A prospective observational study was conducted, enrolling 200 women aged 18 and older with focal breast lesions detected on conventional ultrasound. Participants underwent both strain and shear wave elastography. Lesion characteristics, including stiffness scores and echogenicity, were documented. Histopathological analysis was performed on biopsied lesions to confirm diagnoses. Diagnostic accuracy metrics were calculated for both elastography techniques, including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Results: The average age of participants was 52 ± 10 years. Most lesions were solid (70%), with an average size of 22 mm. Shear wave elastography exhibited higher stiffness scores (average 4.5) than strain elastography (average 3.2). Diagnostic accuracy for benign lesions showed a sensitivity of 85% and specificity of 90% with elastography. For malignant lesions, sensitivity was 90% and specificity was 85%. The ROC curve analysis indicated the superior performance of shear wave elastography with an AUC of 0.85 compared to 0.80 for strain elastography. Conclusion: Ultrasound elastography, particularly shear wave elastography, demonstrates high diagnostic accuracy in characterizing breast lesions, providing valuable support in clinical decision-making.

Computer-Aided Classification of Breast Lesions Based on US RF Time Series Using a Novel Machine Learning Approach.

One of the most promising adjuncts for screening breast cancer is ultrasound (US) radio-frequency (RF) time series. It has the superiority of not requiring any supplementary equipment over other methods. This research aimed to propound a machine learning (ML) approach for automatically classifying benign, probably benign, suspicious, and malignant breast lesions based on the features extracted from the accumulated US RF time series. In this article, 220 data of the aforementioned categories, recorded from 118 patients, were analyzed. The dataset, named RFTSBU, was registered by a SuperSonic Imagine Aixplorer medical/research system equipped with a linear transducer. The regions of interest (ROIs) of the B-mode images were manually selected by an expert radiologist before computing the suggested features. Regarding time, frequency, and time-frequency domains, 291 various features were extracted from each ROI. Finally, the features were classified by a pioneering technique named the reference classification method (RCM). Furthermore, the Lee filter was applied to evaluate the effectiveness of reducing speckle noise on the outcomes. The accuracy of two-class, three-class, and four-class classifications were respectively calculated 98.59 ± 0.71%, 98.13 ± 0.69%, and 96.10 ± 0.66% (considering 10 repetitions) while support vector machine (SVM) and K-nearest neighbor (KNN) classifiers with 5-fold cross-validation were utilized. This article represented the proposed approach, named CCRFML, to distinguish between breast lesions based on registered in vivo RF time series employing an ML framework. The proposed method's impressive level of classification accuracy attests to its capability of effectively assisting medical professionals in the noninvasive differentiation of breast lesions.

Clinical Diagnostic Value of Shear-Wave Elastography in Detecting Malignant Nipple Retraction.

In recent years, the use of shear-wave elastography (SWE) as a diagnostic tool for detecting malignant breast lesions has shown promising results. This study aims to determine the clinical diagnostic value of SWE in detecting malignant nipple retraction. Both US and SWE (Philips EPIQ7 system) were performed for 41 consecutive patients with nipple retraction (56 nipples). The mean, median, and maximum tissue elasticity values (in kilopascals) were determined for each nipple by using SWE. The sensitivity, specificity, and overall accuracy of each measurement was determined by using the surgical pathology results or clinical diagnosis as the gold standard. Of the 56 retracted nipples, 32 were due to benign lesions, and 24 were due to malignant lesions. No significant differences in dimensions or echo features were found between the benign and malignant groups. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the color Doppler flow imaging (CDFI) pattern were 63.89% (23/36), 95% (19/20), 95.83 (23/24), 59.38 (19/32), and 75% (42/56), respectively; the corresponding values for median elasticity on SWE were 88.46% (23/26), 96.67% (29/30), 95.83 (23/24), 90.63 (29/32), and 92.85 (52/56), respectively. The addition of SWE to conventional US could help differentiate benign from malignant lesions associated with nipple retraction.

The Role of Diffusion-Weighted Imaging in Characterizing Benign and Malignant Breast Lesions: A Retrospective Study.

Introduction Diffusion-weighted imaging (DWI) is a promising magnetic resonance imaging (MRI) technique for differentiating between benign and malignant breast lesions. This study set out to assess the diagnostic utility of DWI and apparent diffusion coefficient (ADC) values in the characterization of breast lesions. Materials and methods A retrospective analysis comprised 30 patients with breast lesions who had breast MRI with DWI. The histopathological findings, ADC readings, and conventional MRI features were all analyzed. The receiver operating characteristic (ROC) curve analysis method was utilized to assess the diagnostic accuracy of DWI. Results Out of the 30 lesions, 22 (73.3%) were benign and eight (26.7%) were malignant. Malignant lesions exhibited significantly lower ADC values (p < 0.001) compared to benign lesions. An ADC cutoff value of 1.1 × 10-3mm2/s was optimal for differentiating benign from malignant lesions, yielding 90.81% sensitivity, 91.51% specificity, and 91.5% accuracy. Conclusion Combining DWI with quantitative ADC analysis is a helpful, non-invasive method for the characterization of breast lesions. It shows excellent diagnostic accuracy in identifying benign and malignant lesions, which may cut down on pointless biopsies and help with patient management.

US-based radiomics analysis of different machine learning models for differentiating benign and malignant BI-RADS 4A breast lesions

To investigate and authenticate the effectiveness of various radiomics models in distinguishing between benign and malignant BI-RADS 4A lesions. A total of 936 patients with pathologically confirmed 4A lesions were included in the study (training cohort: n=655; test cohort: n=281). Radiomic features were derived from greyscale US images. Following dimensionality reduction and feature selection, radiomics models were developed using logistic regression (LR), support vector machine (SVM), random forest (RF), eXtreme gradient boosting (XGBoost) and multilayer perceptron (MLP) algorithms. Univariate and multivariable logistic regression analyses were employed to investigate clinical-radiological characteristics and determine variables for creating a clinical model. Five combined models integrating radiomic and clinical parameters were constructed by using each algorithm, and comparison with radiologists' performance was performed. SHapley Additive exPlanations (SHAP) approach was used to elucidate the radiomic model by ranking the significance of features based on their contribution to the evaluation. A total of 1561 radiomic features were extracted. Thirty-six features were deemed significant by dimensionality reduction and selection. The radiomic models showed good performance with AUCs of 0.829-0.945 in training cohort; and 0.805-0.857 in test cohort. The combined model developed by using LR showed the best performance (AUC, training cohort: 0.909; test cohort: 0.905), which is superior to radiologists' performance. Decision curve analysis (DCA) of this combined model indicated better clinical efficacy than clinical and radiomic models. The combined model integrating radiomic and clinical features demonstrated excellent performance in differentiating between benign and malignant 4A lesions. It may offer a non-invasive and efficient approach to aid in clinical decision-making.