Diagnostic performance of 3D automated breast ultrasound (3D-ABUS) in a clinical screening setting-a retrospective study.

OBJECTIVES: To assess the diagnostic performance of 3D automated breast ultrasound (3D-ABUS) in breast cancer screening in a clinical setting. MATERIALS AND METHODS: All patients who had 3D-ABUS between January 2014 and January 2022 for screening were included in this retrospective study. The images were reported by 1 of 6 breast radiologists based on the Breast Imaging Reporting and Data Systems (BI-RADS). The 3D-ABUS was reviewed together with the digital breast tomosynthesis (DBT). Recall rate, biopsy rate, positive predictive value (PPV) and cancer detection yield were calculated. RESULTS: In total, 3616 studies were performed in 1555 women (breast density C/D 95.5% (n = 3455/3616), breast density A/B 4.0% (n = 144/3616), density unknown (0.5% (n = 17/3616)). A total of 259 lesions were detected on 3D-ABUS (87.6% (n = 227/259) masses and 12.4% (n = 32/259) architectural distortions). The recall rate was 5.2% (n = 188/3616) (CI 4.5-6.0%) with only 36.7% (n = 69/188) cases recalled to another date. Moreover, recall declined over time. There were 3.4% (n = 123/3616) biopsies performed, with 52.8% (n = 65/123) biopsies due to an abnormality detected in 3D-ABUS alone. Ten of 65 lesions were malignant, resulting in a positive predictive value (PPV) of 15.4% (n = 10/65) (CI 7.6-26.5%)). The cancer detection yield of 3D-ABUS is 2.77 per 1000 screening tests (CI 1.30-5.1). CONCLUSION: The cancer detection yield of 3D-ABUS in a real clinical screening setting is comparable to the results reported in previous prospective studies, with lower recall and biopsy rates. 3D-ABUS also may be an alternative for screening when mammography is not possible or declined. CLINICAL RELEVANCE STATEMENT: 3D automated breast ultrasound screening performance in a clinical setting is comparable to previous prospective studies, with better recall and biopsy rates. KEY POINTS: • 3D automated breast ultrasound is a reliable and reproducible tool that provides a three-dimensional representation of the breast and allows image visualisation in axial, coronal and sagittal. • The diagnostic performance of 3D automated breast ultrasound in a real clinical setting is comparable to its performance in previously published prospective studies, with improved recall and biopsy rates. • 3D automated breast ultrasound is a useful adjunct to mammography in dense breasts and may be an alternative for screening when mammography is not possible or declined.

A Radiomics Study: Classification of Breast Lesions by Textural Features from Mammography Images.

This study investigates the feasibility of using texture radiomics features extracted from mammography images to distinguish between benign and malignant breast lesions and to classify benign lesions into different categories and determine the best machine learning (ML) model to perform the tasks. Six hundred and twenty-two breast lesions from 200 retrospective patient data were segmented and analysed. Three hundred fifty radiomics features were extracted using the Standardized Environment for Radiomics Analysis (SERA) library, one of the radiomics implementations endorsed by the Image Biomarker Standardisation Initiative (IBSI). The radiomics features and selected patient characteristics were used to train selected machine learning models to classify the breast lesions. A fivefold cross-validation was used to evaluate the performance of the ML models and the top 10 most important features were identified. The random forest (RF) ensemble gave the highest accuracy (89.3%) and positive predictive value (66%) and likelihood ratio of 13.5 in categorising benign and malignant lesions. For the classification of benign lesions, the RF model again gave the highest likelihood ratio of 3.4 compared to the other models. Morphological and textural radiomics features were identified as the top 10 most important features from the random forest models. Patient age was also identified as one of the significant features in the RF model. We concluded that machine learning models trained against texture-based radiomics features and patient features give reasonable performance in differentiating benign versus malignant breast lesions. Our study also demonstrated that the radiomics-based machine learning models were able to emulate the visual assessment of mammography lesions, typically used by radiologists, leading to a better understanding of how the machine learning model arrive at their decision.

Magnetic resonance imaging features of invasive breast cancer association with the tumour stromal ratio.

OBJECTIVE: To assess the association between breast cancer tumour stroma and magnetic resonance imaging (MRI) features. MATERIALS AND METHODS: A total of 84 patients with treatment-naïve invasive breast cancer were enrolled into this retrospective study. The tumour stroma ratio (TSR) was estimated from the amount of tumour stroma in the pathology specimen of the breast tumour. The MRI images of the patients were analysed based on Breast Imaging Reporting and Data Systems (ACR-BIRADS) for qualitative features which include T2- weighted, diffusion-weighted images (DWI) and dynamic contrast-enhanced (DCE) for kinetic features. The mean signal intensity (SI) of Short Tau Inversion Recovery (STIR), with the ratio of STIR of the lesion and pectoralis muscle (L/M ratio) and apparent diffusion coefficient (ADC) value, were measured for the quantitative features. Correlation tests were performed to assess the relationship between TSR and MRI features. RESULTS: There was a significant correlation between the margin of mass, enhancement pattern, and STIR signal intensity of breast cancer and TSR. There were 54.76% (n = 46) in the low stromal group and 45.24% (n = 38) in the high stromal group. A significant association were seen between the margin of the mass and TSR (p = 0.034) between the L/M ratio (p <0.001), and between STIR SI of the lesion and TSR (p<0.001). The median L/M ratio was significantly higher in the high TSR group as compared to the lower TSR group (p < 0.001). CONCLUSION: Breast cancer with high stroma had spiculated margins, lower STIR signal intensity, and a heterogeneous pattern of enhancement. Hence, in this preliminary study, certain MRI features showed a potential to predict TSR.

Clinical usefulness of abbreviated mri protocol in breast cancer detection.

BACKGROUND: The use of breast MRI for screening has increased over the past decade, mostly in women with a high risk of breast cancer. Abbreviated breast MRI (AB-MR) is introduced to make MRI a more accessible screening modality. AB-MR decreases scanning and reporting time and the overall cost of MRI. OBJECTIVE: This study aims to evaluate the diagnostic efficacy of abbreviated MRI protocol in detecting breast cancer in screening and diagnostic populations, using histopathology as the reference standard. MATERIALS AND METHODS: This is a single-centre retrospective cross-sectional study of 134 patients with 198 histologically proven breast lesions who underwent full diagnostic protocol contrast-enhanced breast MRI (FDP-MR) at the University Malaya Medical Centre (UMMC) from 1st January 2018 to 31st December 2019. AB-MR was pre-determined and evaluated with regard to the potential to detect and exclude malignancy from 3 readers of varying radiological experiences. The sensitivity of both AB-MR and FDP-MR were compared using the McNemar test, where both protocols' diagnostic performances were assessed via the receiver operating characteristic (ROC) curve. Inter-observer agreement was analysed using Fleiss Kappa. RESULT: There were 134 patients with 198 lesions. The average age was 50.9 years old (range 27 - 80). A total of 121 (90%) MRIs were performed for diagnostic purposes. Screening accounted for 9.4% of the cases, 55.6% (n=110) lesions were benign, and 44.4% (n=88) were malignant. The commonest benign and malignant lesions were fibrocystic change (27.3%) and invasive ductal carcinoma (78.4%). The mean sensitivity, specificity, positive predictive value, and negative predictive value for AB-MR were 0.96, 0.57, 0.68 and 0.94, respectively. Both AB-MR and FDP-MR showed excellent diagnostic performance with AUC of 0.88 and 0.96, respectively. The general inter-observer agreement of all three readers for AB-MR was substantial (k=0.69), with fair agreement demonstrated between AB-MR and FDP-MR (k=0.36). CONCLUSION: The study shows no evidence that the diagnostic efficacy of AB-MR is inferior to FDP-MR. AB-MR, with high sensitivity, has proven its capability in cancer detection and exclusion, especially for biologically aggressive cancers.

Diagnostic Accuracy of Shear Wave Elastography as an Adjunct Tool in Detecting Axillary Lymph Nodes Metastasis.

OBJECTIVES: This study evaluates the diagnostic performance of shear wave elastography (SWE) in differentiating between benign and axillary lymph node (ALN) metastasis in breast carcinoma. MATERIALS AND METHODS: Breast lesions and axillae of 107 patients were assessed using B-mode ultrasound and SWE. Histopathology was the diagnostic gold standard. RESULTS: In metastatic axillary lymph nodes, qualitative SWE using color patterns had the highest area under curve (AUC) value, followed by B-mode Ultrasound (cortical thickening >3 mm) and quantitative SWE using Emax of 15.2 kPa (AUC of 81.3%, 70.1%, and 61.2%, respectively). Qualitative SWE exhibited better diagnostic performance than the other two parameters, with sensitivity of 96.0% and specificity of 56.1%. Combination of B-mode Ultrasound (using cortical thickness of >3 mm as cut-off point) and qualitative SWE (Color patterns of 2 to 4) showed sensitivity of 71.6%, specificity of 95%, PPV of 96%, NPV of 66.7%, and accuracy of 80.4%. CONCLUSION: Qualitative SWE assessment exhibited higher accuracy compared to quantitative values. Qualitative SWE as an adjunct to B-mode ultrasound can further improve the diagnostic accuracy of metastatic ALN in breast cancer.

Spindle Cell Metaplastic Breast Carcinoma.

INTRODUCTION: Metaplastic breast carcinoma is an uncommon malignancy that constitutes < 5% of all breast cancers. There are 5 subtypes which are spindle cell, squamous cell, carcinosarcoma, matrix-producing and metaplastic with osteoclastic giant cells. Spindle cell carcinoma represents approximately <0.3% of invasive breast carcinomas. It is typically a triple-negative cancer with distinct pathological characteristics, but relatively a non-conclusive on imaging findings. CASE REPORT: An elderly lady presented with an enlarging painful left breast lump for one year. Palpable left breast lump was found on clinical examination. Mammography demonstrated a high density, oval lesion with a partially indistinct margin. Corresponding ultrasound showed a large irregular heterogeneous lesion with solid-cystic areas. Histopathology showed atypical spindle-shaped cells that stained positive for cytokeratins and negative for hormone and human epidermal growth factor receptors, which favoured spindle cell metaplastic carcinoma. Left mastectomy and axillary dissection were performed, and the final diagnosis was consistent with metaplastic spindle cell carcinoma. CONCLUSION: Spindle cell carcinoma of the breast is a rare and aggressive histological type of carcinoma, which may present with benign features on imaging. Tissue diagnosis is essential for prompt diagnosis with multidisciplinary team discussion to guide management and improve patient's outcomes.

Magnetic Resonance Imaging Phenotypes of Breast Cancer Molecular Subtypes: A Systematic Review.

OBJECTIVE: Magnetic resonance imaging (MRI) is the most sensitive imaging modality in detecting breast cancer. The purpose of this systematic review is to investigate the role of human extracted MRI phenotypes in classifying molecular subtypes of breast cancer. METHODS: We performed a literature search of published articles on the application of MRI phenotypic features in invasive breast cancer molecular subtype classifications by radiologists' interpretation on Medline Complete, Pubmed, and Google scholar from 1st January 2000 to 31st March 2021. Of the 1453 literature identified, 42 fulfilled the inclusion criteria. RESULTS: All studies were case-controlled, retrospective study and research-based. The majority of the studies assessed the MRI features using American College of Radiology- Breast Imaging Reporting and Data System (ACR-BIRADS) classification and using dynamic contrast-enhanced (DCE) kinetic features, Apparent Diffusion Coefficient (ADC) values, and T2 sequence. Most studies divided invasive breast cancer into 4 main subtypes, luminal A, luminal B, HER2, and triple-negative (TN) cancers, and used 2 readers. We present a summary of the radiologists' extracted breast MRI phenotypical features and their correlating breast cancer subtypes classifications. The characteristic features are morphology, enhancement kinetics, and T2 signal intensity. We found that the TN subtype has the most distinctive MRI features compared to the other subtypes and luminal A and B have many similar features. CONCLUSION: The MRI features which are predictive of each subtype are the morphology, internal enhancement features, and T2 signal intensity, predominantly between TN and the rest. Radiologists' visual interpretation of some of MRI features may offer insight into the respective invasive breast cancer molecular subtype. However, current evidence are still limited to "suggestive" features instead of a diagnostic standard.  Further research is recommended to explore this potential application, for example, by augmentation of radiologists' visual interpretation by artificial intelligence.

MRI findings suggestive of Alzheimer's disease in patients with primary open angle glaucoma - a single sequence analysis using rapid 3D T1 spoiled gradient echo.

OBJECTIVE: Primary open-angle glaucoma (POAG) is a degenerative optic neuropathy disease which has somewhat similar pathophysiology to Alzheimer's disease (AD). This study aims to determine the presence of medial temporal atrophy and parietal lobe atrophy in patients with POAG compared to normal controls using medial temporal atrophy (MTA) scoring and posterior cortical atrophy (PCA) scoring system on T1 magnetization-prepared rapid gradient-echo. METHODS: 50 POAG patients and 50 normal subjects were recruited and an MRI brain with T1-magnetization-prepared rapid gradient-echo was performed. Medial temporal lobe and parietal lobe atrophy were by MTA and PCA/Koedam scoring. The score of the PCA and MTA were compared between the POAG group and the controls. RESULTS: There was a significant statistical difference between PCA score in POAG and the healthy control group (p-value = 0.026). There is no statistical difference between MTA score in POAG compared to the healthy control group (p-value = 0.58). CONCLUSION: This study suggests a correlation between POAG and PCA score. Potential application of this scoring method in clinical diagnosis and monitoring of POAG patients. ADVANCES IN KNOWLEDGE: The scoring method used in AD may also be applied in the diagnosis and monitoring of POAGMRI brain, specifically rapid volumetric T1 spoiled gradient echo sequence, may be applied in POAG assessment.

Pituitary apoplexy after COVID-19 vaccination: A case report.

Background: Pituitary apoplexy is a rare endocrine emergency, which commonly presents with headache and is occasionally associated with visual disturbances. Prompt diagnosis and treatment can be both life and vision saving. In the emergence of novel coronavirus and global pandemic, rapid development of new vaccines have shown to reduce morbidity and mortality associated with Covid-19. Recognition of rare potential adverse effects of these vaccines including pituitary apoplexy are yet to be reported. A causal link between pituitary apoplexy and COVID-19 vaccination has not been established. Case presentation: We report a case of a 24-year-old woman who presented with progressively worsening headache soon after completing her COVID-19 vaccination. Imaging showed pituitary apoplexy with an underlying pituitary mass. In view of the age and the typical presentation of severe headache, pituitary hypophysitis was considered, despite the absence of the almost pathognomonic feature of a thickened pituitary stalk in the initial imaging. In the context that the headache had started shortly after the administration of the second dose of COVID-19 vaccine, this potentially could have been the trigger for the occurrence of pituitary apoplexy. Conclusion: Although the pathophysiology is not entirely clear and no direct link could be ascertained, our patient may have developed an exaggerated immunological response after the vaccine, with a possible pituitary hypophysitis leading to a pituitary apoplexy.

Artificial intelligence, BI-RADS evaluation and morphometry: A novel combination to diagnose breast cancer using ultrasonography, results from multi-center cohorts.

PURPOSE: To develop and validate a machine learning (ML) model for the classification of breast lesions on ultrasound images. METHOD: In the present study, three separate data cohorts containing 1288 breast lesions from three countries (Malaysia, Iran, and Turkey) were utilized for MLmodel development and external validation. The model was trained on ultrasound images of 725 breast lesions, and validation was done separately on the remaining data. An expert radiologist and a radiology resident classified the lesions based on the BI-RADS lexicon. Thirteen morphometric features were selected from a contour of the lesion and underwent a three-step feature selection process. Five features were chosen to be fed into the model separately and combined with the imaging signs mentioned in the BI-RADS reference guide. A support vector classifier was trained and optimized. RESULTS: The diagnostic profile of the model with various input data was compared to the expert radiologist and radiology resident. The agreement of each approach with histopathologic specimens was also determined. Based on BI-RADS and morphometric features, the model achieved an area under the receiver operating characteristic (ROC) curve (AUC) of 0.885, which is higher than the expert radiologist and radiology resident performances with AUC of 0.814 and 0.632, respectively in all cohorts. DeLong's test also showed that the AUC of the ML protocol was significantly different from that of the expert radiologist (ΔAUCs = 0.071, 95%CI: (0.056, 0.086), P = 0.005). CONCLUSIONS: These results support the possible role of morphometric features in enhancing the already well-excepted classification schemes.

Evaluation of automated volumetric breast density software in comparison with visual assessments in an Asian population: A retrospective observational study.

This study aims to compare Quantra, as an automated volumetric breast density (Vbd) tool, with visual assessment according to ACR BI-RADS density categories and to determine its potential usage in clinical practice.Five hundred randomly selected screening and diagnostic mammograms were included in this retrospective study. Three radiologists independently assigned qualitative ACR BI-RADS density categories to the mammograms. Quantra automatically calculates the volumetric density data into the system. The readers were blinded to the Quantra and other readers assessment. Inter-reader agreement and agreement between Quantra and each reader were tested. Region under the curve (ROC) analysis was performed to obtain the cut-off value to separate dense from a non-dense breast. Results with P value <.05 was taken as significant.There were 40.4% Chinese, 27% Malays, 19% Indian and 3.6% represent other ethnicities. The mean age of the patients was 57. 15%, 45.6%, 30.4%, and 9% of patients fall under BI-RADS A, B, C and D density category respectively. Fair agreement with Kappa (κ) value: 0.49, 0.38, and 0.30 were seen for Reader 1, 2 and 3 versus Quantra. Moderate agreement with κ value: 0.63, 0.64, 0.51 was seen when the data were dichotomized (density A and B to "non-dense", C and D to "dense"). The cut-off Vbd value was 13.5% to stratify dense from non-dense breasts with a sensitivity of 86.2% and specificity of 83.1% (AUC 91.4%; confidence interval: 88.8, 94.1).Quantra showed moderate agreement with radiologists visual assessment. Hence, this study adds to the available evidence to support the potential use of Quantra as an adjunct tool for breast density assessment in routine clinical practice in the Asian population. We found 13.5% is the best cut-off value to stratify dense to non-dense breasts in our study population. Its application will provide an objective, consistent and reproducible results as well as aiding clinical decision-making on the need for supplementary breast ultrasound in our screening population.

Diagnostic accuracy of digital breast tomosynthesis in combination with 2D mammography for the characterisation of mammographic abnormalities.

This study aims to assess the diagnostic accuracy of digital breast tomosynthesis in combination with full field digital mammography (DBT + FFDM) in the charaterisation of Breast Imaging-reporting and Data System (BI-RADS) category 3, 4 and 5 lesions. Retrospective cross-sectional study of 390 patients with BI-RADS 3, 4 and 5 mammography with available histopathology examination results were recruited from in a single center of a multi-ethnic Asian population. 2 readers independently reported the FFDM and DBT images and classified lesions detected (mass, calcifications, asymmetric density and architectural distortion) based on American College of Radiology-BI-RADS lexicon. Of the 390 patients recruited, 182 malignancies were reported. Positive predictive value (PPV) of cancer was 46.7%. The PPV in BI-RADS 4a, 4b, 4c and 5 were 6.0%, 38.3%, 68.9%, and 93.1%, respectively. Among all the cancers, 76% presented as masses, 4% as calcifications and 20% as asymmetry. An additional of 4% of cancers were detected on ultrasound. The sensitivity, specificity, PPV and NPV of mass lesions detected on DBT + FFDM were 93.8%, 85.1%, 88.8% and 91.5%, respectively. The PPV for calcification is 61.6% and asymmetry is 60.7%. 81.6% of cancer detected were invasive and 13.3% were in-situ type. Our study showed that DBT is proven to be an effective tool in the diagnosis and characterization of breast lesions and supports the current body of literature that states that integrating DBT to FFDM allows good characterization of breast lesions and accurate diagnosis of cancer.