A deep learning-based method for the detection and segmentation of breast masses in ultrasound images.

Objective.Automated detection and segmentation of breast masses in ultrasound images are critical for breast cancer diagnosis, but remain challenging due to limited image quality and complex breast tissues. This study aims to develop a deep learning-based method that enables accurate breast mass detection and segmentation in ultrasound images.Approach.A novel convolutional neural network-based framework that combines the You Only Look Once (YOLO) v5 network and the Global-Local (GOLO) strategy was developed. First, YOLOv5 was applied to locate the mass regions of interest (ROIs). Second, a Global Local-Connected Multi-Scale Selection (GOLO-CMSS) network was developed to segment the masses. The GOLO-CMSS operated on both the entire images globally and mass ROIs locally, and then integrated the two branches for a final segmentation output. Particularly, in global branch, CMSS applied Multi-Scale Selection (MSS) modules to automatically adjust the receptive fields, and Multi-Input (MLI) modules to enable fusion of shallow and deep features at different resolutions. The USTC dataset containing 28 477 breast ultrasound images was collected for training and test. The proposed method was also tested on three public datasets, UDIAT, BUSI and TUH. The segmentation performance of GOLO-CMSS was compared with other networks and three experienced radiologists.Main results.YOLOv5 outperformed other detection models with average precisions of 99.41%, 95.15%, 93.69% and 96.42% on the USTC, UDIAT, BUSI and TUH datasets, respectively. The proposed GOLO-CMSS showed superior segmentation performance over other state-of-the-art networks, with Dice similarity coefficients (DSCs) of 93.19%, 88.56%, 87.58% and 90.37% on the USTC, UDIAT, BUSI and TUH datasets, respectively. The mean DSC between GOLO-CMSS and each radiologist was significantly better than that between radiologists (p< 0.001).Significance.Our proposed method can accurately detect and segment breast masses with a decent performance comparable to radiologists, highlighting its great potential for clinical implementation in breast ultrasound examination.

Signaling events at TMEM doorways provide potential targets for inhibiting breast cancer dissemination.

Tumor cell intravasation is essential for metastatic dissemination, but its exact mechanism is incompletely understood. We have previously shown that in breast cancer, the direct and stable association of a tumor cell expressing Mena, a Tie2hi/VEGFhi macrophage, and a vascular endothelial cell, creates an intravasation portal, called a "tumor microenvironment of metastasis" (TMEM) doorway, for tumor cell intravasation, leading to dissemination to distant sites. The density of TMEM doorways, also called TMEM doorway score, is a clinically validated prognostic marker of distant metastasis in breast cancer patients. Although we know that tumor cells utilize TMEM doorway-associated transient vascular openings to intravasate, the precise signaling mechanisms involved in TMEM doorway function are only partially understood. Using two mouse models of breast cancer and an in vitro assay of intravasation, we report that CSF-1 secreted by the TMEM doorway tumor cell stimulates local secretion of VEGF-A from the Tie2hi TMEM doorway macrophage, leading to the dissociation of endothelial junctions between TMEM doorway associated endothelial cells, supporting tumor cell intravasation. Acute blockade of CSF-1R signaling decreases macrophage VEGF-A secretion as well as TMEM doorway-associated vascular opening, tumor cell trans-endothelial migration, and dissemination. These new insights into signaling events regulating TMEM doorway function should be explored further as treatment strategies for metastatic disease.

Tracking Tumor Cell Dissemination from Lung Metastases Using Photoconversion.

Metastasis - the systemic spread of cancer - is the leading cause of cancer-related deaths. Although metastasis is commonly thought of as a unidirectional process wherein cells from the primary tumor disseminate and seed metastases, tumor cells in existing metastases can also redisseminate and give rise to new lesions in tertiary sites in a process known as "metastasis-from-metastases" or "metastasis-to-metastasis seeding." Metastasis-to-metastasis seeding may increase the metastatic burden and decrease the patient's quality of life and survival. Therefore, understanding the processes behind this phenomenon is crucial to refining treatment strategies for patients with metastatic cancer. Little is known about metastasis-to-metastasis seeding, due in part to logistical and technological limitations. Studies on metastasis-to-metastasis seeding rely primarily on sequencing methods, which may not be practical for researchers studying the exact timing of metastasis-to-metastasis seeding events or what promotes or prevents them. This highlights the lack of methodologies that facilitate the study of metastasis-to-metastasis seeding. To address this, we have developed - and describe herein - a murine surgical protocol for the selective photoconversion of lung metastases, allowing specific marking and fate tracking of tumor cells redisseminating from the lung to tertiary sites. To our knowledge, this is the only method for studying tumor cell redissemination and metastasis-to-metastasis seeding from the lungs that does not require genomic analysis.

Racial disparity in tumor microenvironment and distant recurrence in residual breast cancer after neoadjuvant chemotherapy.

Black, compared to white, women with residual estrogen receptor-positive (ER+) breast cancer after neoadjuvant chemotherapy (NAC) have worse distant recurrence-free survival (DRFS). Such racial disparity may be due to difference in density of portals for systemic cancer cell dissemination, called TMEM doorways, and pro-metastatic tumor microenvironment (TME). Here, we evaluate residual cancer specimens after NAC from 96 Black and 87 white women. TMEM doorways are visualized by triple immunohistochemistry, and cancer stem cells by immunofluorescence for SOX9. The correlation between TMEM doorway score and pro-metastatic TME parameters with DRFS is examined using log-rank and multivariate Cox regression. Black, compared to white, patients are more likely to develop distant recurrence (49% vs 34.5%, p = 0.07), receive mastectomy (69.8% vs 54%, p = 0.04), and have higher grade tumors (p = 0.002). Tumors from Black patients have higher TMEM doorway and macrophages density overall (p = 0.002; p = 0.002, respectively) and in the ER+/HER2- (p = 0.02; p = 0.02, respectively), but not in the triple negative disease. Furthermore, high TMEM doorway score is associated with worse DRFS. TMEM doorway score is an independent prognostic factor in the entire study population (HR, 2.02; 95%CI, 1.18-3.46; p = 0.01), with a strong trend in ER+/HER2- disease (HR, 2.38; 95%CI, 0.96-5.95; p = 0.06). SOX9 expression is not associated with racial disparity in TME or outcome. In conclusion, higher TMEM doorway density in residual breast cancer after NAC is associated with higher distant recurrence risk, and Black patients are associated with higher TMEM doorway density, suggesting that TMEM doorway density may contribute to racial disparities in breast cancer.

An integrated ultrasound imaging and abdominal compression device for respiratory motion management in radiation therapy.

BACKGROUND: Radiotherapy to tumors in the abdomen is challenging because of the significant organ movement and tissue deformation caused by respiration. PURPOSE: A motion management strategy that integrated ultrasound (US) imaging with abdominal compression was developed and evaluated, where US was used to real-time monitor organ motion after abdominal compression. METHODS: A device that combined a US imaging system and an abdominal compression plate (ACP) was developed. Twenty-one healthy volunteers were involved to evaluate the motion management efficacy. Each volunteer was immobilized on a flat bench by the device. Abdominal US data were successively collected with and without ACP compression, and experiments were repeated three times to verify the imaging reproducibility. A template matching algorithm based on normalized cross-correlation was implemented to track the targets (vessels in the liver, pancreas, and stomach) automatically. The matching algorithm was validated by comparing with the manual references. Automatic tracking was judged as failed if the center-of-mass difference from manual tracking was beyond a failure threshold. Based on the locations obtained through the template matching algorithm, the motion correlation between liver and pancreas/stomach was investigated using the Pearson correlation test. Paired Student's t-test was used to analyze the difference between the results without and with ACP compression. RESULTS: The liver motion amplitude over all 21 volunteers was significantly (p < 0.001) reduced from 14.9 ± 5.5/3.4 ± 1.8 mm in superior-inferior (SI)/anterior-posterior (AP) direction before ACP compression to 7.3 ± 1.5/1.6 ± 0.7 mm after ACP compression. The mean liver motion standard deviation before compression was on average 2.8/1.4 mm in SI/AP direction and was significantly (p < 0.001) reduced to 0.9/0.4 mm after compression. The failure rates of automatic tracking for liver, pancreas, and stomach were reduced for failure thresholds of 1-5 mm after applying ACP. The Pearson correlation coefficients between liver and pancreas/stomach were 0.98/0.97 without ACP and 0.96/0.94 with ACP in the SI direction and were 0.68/0.68 and 0.43/0.42 in the AP direction. The motion prediction errors for pancreas/stomach with ACP have significantly (p < 0.001) reduced to 0.45 ± 0.36/0.52 ± 0.43 mm from 0.69 ± 0.56/0.71 ± 0.66 mm without ACP in the SI direction, and to 0.38 ± 0.33/0.39 ± 0.27 mm from 0.44 ± 0.35/0.61 ± 0.59 mm in the AP direction. CONCLUSIONS: The proposed strategy that combines real-time US imaging and abdominal compression has the potential to reduce the abdominal organ motion while improving both target tracking reliability and motion reproducibility. Furthermore, the observed correlation between liver and pancreas/stomach motion indicates the possibility of indirect pancreas/stomach tracking using liver markers as tracking surrogates. The strategy is expected to provide an alternative for respiratory motion management in the radiation treatment of abdominal tumors.

SWIP-a stabilized window for intravital imaging of the murine pancreas.

Pancreatitis and pancreatic ductal adenocarcinoma (PDAC) are grave illnesses with high levels of morbidity and mortality. Intravital imaging (IVI) is a powerful technique for visualizing physiological processes in both health and disease. However, the application of IVI to the murine pancreas presents significant challenges, as it is a deep, compliant, visceral organ that is difficult to access, easily damaged and susceptible to motion artefacts. Existing imaging windows for stabilizing the pancreas during IVI have unfortunately shown poor stability for time-lapsed imaging on the minutes to hours scale, or are unable to accommodate both the healthy and tumour-bearing pancreata. To address these issues, we developed an improved stabilized window for intravital imaging of the pancreas (SWIP), which can be applied to not only the healthy pancreas but also to solid tumours like PDAC. Here, we validate the SWIP and use it to visualize a variety of processes for the first time, including (1) single-cell dynamics within the healthy pancreas, (2) transformation from healthy pancreas to acute pancreatitis induced by cerulein, and (3) the physiology of PDAC in both autochthonous and orthotopically injected models. SWIP can not only improve the imaging stability but also expand the application of IVI in both benign and malignant pancreas diseases.

Analysis of Risk Factors Associated With Central Lymph Node Metastasis in Papillary Thyroid Carcinoma With cT1N0 Stage.

Aim: Annual T1 stage papillary thyroid carcinoma (PTC) incidence rates continue to rise, yet the optimal treatment for this cancer type remains controversial. Central lymph node metastasis (CLNM) is a critical determinant in the context of treatment decision-making. While several prior studies have evaluated patients with clinica l T1a(cT1a) stage PTC, there have been fewer analyses of clinical T1b(cT1b) disease to date. The present study was thus formulated to explore predictors of CLNM in patients with cT1a and cT1b stage PTC. Methods: A retrospective analysis of data including clinicopathological characteristics and BRAFV600E mutation status was conducted for 452 PTC patients undergoing surgical treatment. Logistic univariate and multivariate analyses were performed to identify risk factors associated with CLNM in particular patients' characteristics and the accuracy of the established logistic regression models was evaluated using the R software platform. Results: Respective CLNM incidence rates in cT1a and cT1b disease were 39.39% and 67.21%. Factors associated with a higher risk of CLNM among PTC(cT1a) patients included male sex, young age, tumor size, contact with capsule, and multifocality as determined through comparisons of the area under the curve for logistic regression models. Whereas male sex and age were associated with CLNM risk in PTC(cT1b) patients in univariate and multivariate analyses, age was the only risk factor associated with CLNM incidence among women with PTC(cT1b). Conclusion: Predictors of CLNM differ between PTC patients with cT1a and cT1b stage disease, and a comprehensive assessment of these risk factors should thus be conducted when designing individualized treatment regimens for PTC patients.

Evaluation of Tissue Stiffness Around Lesions by Sound Touch Shear Wave Elastography in Breast Malignancy Diagnosis.

The aim of the study described here was to assess the evaluation of tissue stiffness around lesions by sound touch shear wave elastography (STE) in breast malignancy diagnosis. This was an institutional ethics committee-approved, single-center study. A total of 90 women with breast masses examined with conventional ultrasound and STE were eligible for enrollment from December 2020 to July 2021. The maximum and mean elastic values of masses, Emax and Emean, were determined. Shell function was used to measure the maximum and mean elastic values of tissues around masses in annular shells 0.5, 1.0, 1.5 and 2.0 mm wide, recorded as corresponding Emax-shell and Emean-shell. All parameters were analyzed and compared with histopathologic results. Receiver operating characteristic curves were constructed to assess diagnostic performance. Logistic regression analysis was conducted to determine the best diagnostic model. Collagen fiber content of tissues around breast lesions was evaluated using Masson staining and ImageJ software. Ninety women with breast masses were included in this study; 50 had benign (mean diameter 15.84 ± 4.39 mm) and 40 had malignant (mean diameter 17.40 ± 5.42 mm) masses. The diagnostic value of Emax-shell-2.0 was the highest (area under the curve = 0.930) with a sensitivity of 87.5% and specificity of 88%. According to stepwise logistic regression analysis, Emax-shell-2.0 and age were independent predictors of malignancy. Emax-shell-2.0 was also found to be highly correlated with the collagen fiber content of tissue in the malignant group (r = 0.877). Tissue stiffness around lesions measured by STE is a useful metric in identifying malignant breast masses by reflecting collagen fiber content, and Emax-shell-2.0 performs best.

Combining TMEM Doorway Score and MenaCalc Score Improves the Prediction of Distant Recurrence Risk in HR+/HER2- Breast Cancer Patients.

PURPOSE: to develop several digital pathology-based machine vision algorithms for combining TMEM and MenaCalc scores and determine if a combination of these biomarkers improves the ability to predict development of distant metastasis over and above that of either biomarker alone. METHODS: This retrospective study included a subset of 130 patients (65 patients with no recurrence and 65 patients with a recurrence at 5 years) from the Calgary Tamoxifen cohort of breast cancer patients. Patients had confirmed invasive breast cancer and received adjuvant tamoxifen therapy. Of the 130 patients, 86 cases were suitable for analysis in this study. Sequential sections of formalin-fixed paraffin-embedded patient samples were stained for TMEM doorways (immunohistochemistry triple staining) and MenaCalc (immunofluorescence staining). Stained sections were imaged, aligned, and then scored for TMEM doorways and MenaCalc. Different ways of combining TMEM doorway and MenaCalc scores were evaluated and compared to identify the best performing combined marker by using the restricted mean survival time (RMST) difference method. RESULTS: the best performing combined marker gave an RMST difference of 5.27 years (95% CI: 1.71-8.37), compared to 3.56 years (95% CI: 0.95-6.1) for the associated standalone TMEM doorway analysis and 2.94 years (95% CI: 0.25-5.87) for the associated standalone MenaCalc analysis. CONCLUSIONS: combining TMEM doorway and MenaCalc scores as a new biomarker improves prognostication over that observed with TMEM doorway or MenaCalc Score alone in this cohort of 86 patients.

Evaluation of Ultrasound Elastography Combined With Chi-Square Automatic Interactive Detector in Reducing Unnecessary Fine-Needle Aspiration on TIRADS 4 Thyroid Nodules.

BACKGROUND: Conventional ultrasound diagnosis of thyroid nodules (TNs) had a high false-positive rate, resulting in many unnecessary fine-needle aspirations (FNAs). OBJECTIVE: This study aimed to establish a simple algorithm to reduce unnecessary FNA on TIRADS 4 TNs using different quantitative parameters of ultrasonic elasticity and chi-square automatic interactive detector (CHAID) method. METHODS: From January 2020 to May 2021, 432 TNs were included in the study, which were confirmed by FNA or surgical pathology. Each TN was examined using conventional ultrasound, sound touch elastography, and Shell measurement function. The quantitative parameters E and E shell were recorded, and the E shell/E values were calculated for each TN. The diagnostic performance of the quantitative parameters was evaluated using the receiver operating characteristic curves. The CHAID was used to classify and analyze the quantitative parameters, and the prediction model was established. RESULTS: A total of 226 TNs were malignant and 206 were benign. E shell and E shell/E ratio were included in the classification algorithm, which showed a depth of two ramifications (E shell/E ≤ 0.988 or 0.988-1.043 or >1.043; if E shell/E ≤ 0.988, then E shell ≤ 64.0 or 64.0-74.0 or >74.0; if E shell/E = 0.988-1.043, then E shell ≤ 66.0 or > 66.0; if E shell/E >1.043, then E shell ≤ 69.0 or >69.0). The unnecessary FNAs could have been avoided in 57.3% of the cases using this algorithm. CONCLUSION: The prediction model using quantitative parameters had high diagnostic performance; it could quickly distinguish benign lesions and avoid subjective influence to some extent.

Primary tumor associated macrophages activate programs of invasion and dormancy in disseminating tumor cells.

Metastases are initiated by disseminated tumor cells (DTCs) that colonize distant organs. Growing evidence suggests that the microenvironment of the primary tumor primes DTCs for dormant or proliferative fates. However, the manner in which this occurs remains poorly understood. Here, using the Window for High-Resolution Intravital Imaging of the Lung (WHRIL), we study the live lung longitudinally and follow the fate of individual DTCs that spontaneously disseminate from orthotopic breast tumors. We find that spontaneously DTCs have increased levels of retention, increased speed of extravasation, and greater survival after extravasation, compared to experimentally metastasized tumor cells. Detailed analysis reveals that a subset of macrophages within the primary tumor induces a pro-dissemination and pro-dormancy DTC phenotype. Our work provides insight into how specific primary tumor microenvironments prime a subpopulation of cells for expression of proteins associated with dissemination and dormancy.

Live tumor imaging shows macrophage induction and TMEM-mediated enrichment of cancer stem cells during metastatic dissemination.

Cancer stem cells (CSCs) play an important role during metastasis, but the dynamic behavior and induction mechanisms of CSCs are not well understood. Here, we employ high-resolution intravital microscopy using a CSC biosensor to directly observe CSCs in live mice with mammary tumors. CSCs display the slow-migratory, invadopod-rich phenotype that is the hallmark of disseminating tumor cells. CSCs are enriched near macrophages, particularly near macrophage-containing intravasation sites called Tumor Microenvironment of Metastasis (TMEM) doorways. Substantial enrichment of CSCs occurs on association with TMEM doorways, contributing to the finding that CSCs represent >60% of circulating tumor cells. Mechanistically, stemness is induced in non-stem cancer cells upon their direct contact with macrophages via Notch-Jagged signaling. In breast cancers from patients, the density of TMEM doorways correlates with the proportion of cancer cells expressing stem cell markers, indicating that in human breast cancer TMEM doorways are not only cancer cell intravasation portals but also CSC programming sites.

Evaluation of the Perinodular Stiffness Potentially Predicts the Malignancy of Thyroid Nodules.

OBJECTIVES: To evaluate the surrounding tissue stiffness measured by sound touch elastography for differential diagnosis of thyroid nodules (TNs). METHODS: Thirty-nine benign and 90 malignant TNs were included in this study. The conventional ultrasound features, the maximum Young modulus value of the stiffness of the TNs (recorded as E), and the stiffness of the 0.5-, 1.0-, 1.5-, and 2.0-mm perinodular regions of the TNs (recorded as Eshell0.5 , Eshell1.0 , Eshell1.5 , and Eshell2.0 , respectively) were prospectively analyzed and compared to histopathologic results. The abundance of collagen fibers at various widths in the perinodular regions of the TNs was evaluated by Masson staining and ImageJ software (National Institutes of Health, Bethesda, MD). The fibrous structures in the perinodular regions of the TNs were classified. RESULTS: The various Eshell values of malignant TNs were significantly higher than those of benign TNs (P < .001 for all). Eshell0.5 correlated highly with E in the malignant TNs and in all samples (r = 0.722 and 0.772; P < .001 for both). Eshell2 yielded the highest area under the receiving operating characteristic curve value (0.96) for the differential diagnosis of TNs. The abundance of collagen fibers in the 2-mm perinodular region of the TNs was closely correlated with Eshell2 in the malignant TNs and in all samples (r =0.729 and 0.867; P < .001). The Eshell2 values for different levels of disorder of the tissue surrounding TNs were significantly different (P < .01 for all). CONCLUSIONS: Perinodular stiffness measured by sound touch elastography improved the diagnostic accuracy in TNs.

Feasibility of Point Shear Wave Elastography for Evaluating Diabetic Peripheral Neuropathy.

OBJECTIVES: In the mode of Virtual Touch quantification (Siemens AG, Erlangen, Germany), point shear wave elastography (p-SWE) is widely used for noninvasive assessments of tissue stiffness, which may be useful in the evaluation of diabetic peripheral neuropathy (DPN). METHODS: Thirty patients with type 2 diabetes and 20 control participants (7 with myoma of the uterus and 13 with kidney stones) were enrolled in this prospective study. The 30 patients were further divided into patients with DPN and patients without DPN. Conventional ultrasound examinations and p-SWE were used to examine the tibial nerve in the popliteal fossa. RESULTS: Tibial nerve stiffness values in the overall patient group, patients with DPN, and patients without DPN were all significantly higher than in the control group (P < .05). The cutoff value of p-SWE for assessing DPN was 2.60 m/s; at that threshold, sensitivity was 63.33%, and specificity was 92.50%. CONCLUSIONS: Point SWE was useful for the noninvasive assessment of DPN and had high specificity. The increased stiffness in patients without DPN indicated that the tibial nerve might be affected by diabetes.