Search

Medical ultrasound imaging technology is currently the preferred method for early diagnosis of thyroid nodules. Radiologists' analysis of ultrasound images is highly dependent on their clinical experience and is susceptible to intra- and inter-observer variability. Although end-to-end deep learning technique can address these limitations, the difficulty of acquiring annotated medical image makes it very challenging. Transfer learning can alleviate the problems, but the large gap between source and target domain will lead to negative transfer. In this paper, a novel transfer learning method with distant domain high-level feature fusion (DHFF) model is proposed. It reduces the distribution distance between the source domain and the target domain while maintaining the characteristics of respective domains, which can avoid excessive feature fusion while enabling the model to learn more valuable transfer knowledge. The DHFF is validated by multiple public source and private target datasets in experiments. The results show that the classification accuracy of DHFF is up to 88.92% with thyroid ultrasound auxiliary source domains, which is up to 8% higher than existing transfer and distant transfer algorithms.

Renal angiomyolipoma (RAML), a rare benign mesenchymal tumor, consists of variable proportions of thick-walled blood vessels admixed with adipose tissue and bundles of smooth muscle. Rarely, RAML angiomyolipoma may be associated with pseudoaneurysm, spontaneous rupture and hemorrhage. we herein present an exceptional case of fat-poor RAML containing a large pseudoaneurysm in a 48-yearold woman with known pulmonary lymphangioleiomyomatosis (PLAM), who was admitted to hospital due to right flank pain for about two months. To avoid risk of rupture of the pseudoaneurysms, transcatheter arterial embolization (TAE) procedure was performed, at the same time taking into account of the possibility of malignancy in such a large neoplasm, right radical nephrectomy was taken five days later. This case was benign and the 1-year follow-up ultrasonic examination showed no recurrence. This is the first report that the patient had a sporadic giant RAML with pseudoaneurysm which was not associated with the TSC, and to our knowledge, such cases have not been explored in literature. This case reminds us that giant sporadic RAML with pseudoaneurysm, although rare, does exist, and we should improve the understanding of the neoplasm. Early detection and therapeutic interventions of pseudoaneurysm in an AML patient can reduce serious complications.

To examine the association of alcohol consumption with the presence and development of ultrasound (US)-detected tophi and subcutaneous tophi (sub-tophi) in a Chinese gout population.A total of 554 patients with gout who underwent US and physical examination on the most frequently involved joints in gout were included. Multivariable analysis was performed to assess the associations of the duration, quantity, and type of alcohol consumption with the presence, size, and number of US-detected tophi and sub-tophi.Compared with non-drinkers, excessive drinkers (70 g/week), long-term drinkers (≥10 years) and spirits drinkers had a greater proportion, size, and number of US-detected tophi and sub-tophi (all P 0.05). After adjusting for confounders, excessive drinking ( 70 g/week) (odds ratio [OR]: 1.79; OR: 2.00), long-term alcohol consumption (≥ 10 years) (OR: 1.96; OR: 2.17), and spirits consumption of (OR: 1.81; OR: 2.10) were significantly associated with the presence of US-detected tophi and sub-tophi (all P 0.05), with the highest OR among the identified risk factors. Among patients who already have US-detected tophi or sub-tophi, moderate drinking (≤ 70 g/week) was associated with larger or multiple tophi (all P 0.05).A longer duration and higher quantity of alcohol consumption as well as spirits consumption are predictors for the development of US-detected tophi and sub-tophi in patients with gout. Among people who have US-detected tophi and sub-tophi, weekly alcohol consumption leads to the development of tophi regardless of amount consumed. This article is protected by copyright. All rights reserved.

To determine the clinical value of ultrasonic features, especially extrathyroidal extension (ETE), in the prediction of PTC recurrence.A total of 863 patients with PTC confirmed by pathological examinations from January 2012 to August 2018 were selected in this study, including 59 cases of recurrence. The Cox-proportional hazards regression analysis and Kaplan-Meier method were adopted to determine the relationship between the variables and recurrence free survival (RFS).The recurrence rate of PTC is 6.8 %. Tumor maximum diameter, margin, multifocality, microcalcifications, ETE and preoperative lymph node metastasis were valuable predictive factors in univariate survival analysis. Tumor larger than 20 mm, multifocality and lateral cervical lymph node metastasis were independent risk factors for PTC recurrence, and lymph node metastasis has the highest hazard ratio (HR). Preoperative lateral cervical lymph node metastasis was more often found in the gross and extensive ETE groups. Microscopic ETE has little value in predicting PTC recurrence and has no correlation with preoperative cervical lymph node metastasis.Tumor maximum diameter20 mm, multifocality and lateral cervical lymph node metastasis were independent risk factors for PTC recurrence. Preoperative lateral cervical lymph nodes should be carefully examined when gross ETE and extensive ETE were detected. Microscopic ETE has no impact on preoperative cervical lymph node metastasis or tumor recurrence.

Background This study aimed to examine the performance of the four risk of malignancy index (RMI) in discriminating borderline ovarian tumors (BOTs) and benign ovarian masses in daily clinical practice. Methods A total of 162 women with BOTs and 379 women with benign ovarian tumors diagnosed at the Second Affiliated Hospital of Harbin Medical University from January 2012 to December 2016 were enrolled in this retrospective study. Also, we classified these patients into serous borderline ovarian tumor (SBOT) and mucinous borderline ovarian tumor (MBOT) subgroup. Preoperative ultrasound findings, cancer antigen 125 (CA125) and menopausal status were reviewed. The area under the curve (AUC) of receiver operator characteristic curves (ROC) and performance indices of RMI I, RMI II, RMI III and RMI IV were calculated and compared for discrimination between benign ovarian tumors and BOTs. Results RMI I had the highest AUC (0.825, 95% CI: 0.790–0.856) among the four RMIs in BOTs group. Similar results were found in SBOT (0.839, 95% CI: 0.804–0.871) and MBOT (0.791, 95% CI: 0.749–0.829) subgroups. RMI I had the highest specificity among the BOTs group (87.6, 95% CI: 83.9–90.7%), SBOT (87.6, 95% CI: 83.9–90.7%) and MBOT group (87.6, 95% CI: 83.9–90.7%). RMI II scored the highest overall in terms of sensitivity among the BOTs group (69.75, 95% CI: 62.1–76.7%), SBOT (74.34, 95% CI: 65.3–82.1%) and MBOT (59.18, 95% CI: 44.2–73.0%) group. Conclusion Compared to other RMIs, RMI I was the best-performed method for differentiation of BOTs from benign ovarian tumors. At the same time, RMI I also performed best in the discrimination SBOT from benign ovarian tumors.

Breast cancer investigation is of great significance, and developing tumor detection methodologies is a critical need. However, it is challenging for breast cancer detection using breast ultrasound (BUS) images due to the complicated breast structure and poor quality of the images. This paper proposes a novel tumor saliency estimation (TSE) model guided by enriched breast anatomy knowledge to localize the tumor. First, the breast anatomy layers are generated by a deep neural network. Then we refine the layers by integrating a non-semantic breast anatomy model to solve the problems of incomplete mammary layers. Meanwhile, a new background map generation method weighted by the semantic probability and spatial distance is proposed to improve the performance. The experiment demonstrates that the proposed method with the new background map outperforms four state-of-the-art TSE models with an increasing 10% of $F_{measure}$ on the public BUS dataset.

Tumor saliency estimation aims to localize tumors by modeling the visual stimuli in medical images. However, it is a challenging task for breast ultrasound (BUS) image due to the complicated anatomic structure of the breast and poor image quality; and existing saliency estimation approaches only model the generic visual stimuli, e.g., local and global contrast, location, and feature correlation, and achieve poor performance for tumor saliency estimation. In this paper, we propose a novel optimization model to estimate tumor saliency by utilizing breast anatomy. First, we model breast anatomy and decompose breast ultrasound image into layers using Neutro-Connectedness; then utilize the layers to generate the foreground and background maps; and finally propose a novel objective function to estimate the tumor saliency by integrating the foreground map, background map, adaptive center bias, and region-based correlation cues. The extensive experiments demonstrate that the proposed approach obtains more accurate foreground and background maps with breast anatomy; especially, for the images having large or small tumors. Meanwhile, the new objective function can handle the images without tumors. The newly proposed method achieves state-of-the-art performance comparing to eight tumor saliency estimation approaches using two BUS datasets.

OBJECTIVE The aim of our study was to evaluate the role of preoperative US, CEUS, and 99mTc-MIBI scanning with SPECT/CT in localizing diseased parathyroid glands in cases of refractory secondary hyperparathyroidism (SHPT). MATERIAL AND METHODS Using pathological results as the gold standard, we compared the operative findings with the preoperative localization of each modality in 73 nodules and evaluated the accuracy, and sensitivity of each modality and combinations of the four modalities. RESULTS The sensitivity of US, CEUS, 99mTc-MIBI and SPECT/CT was 98.59%, 94.37%, 50.70% and 78.87%, respectively. US had the highest sensitivity of the four imaging methods and the diagnostic sensitivity of US and CEUS was superior to that of 99mTc-MIBI (p

Medical ultrasound imaging is widely used in clinical diagnosis because of its non-invasive, convenient and quick characteristics. However, due to its low image contrast, multiple artifacts, noise and lack of paired high-resolution and low-resolution image data sets, the task of super-resolution reconstruction of medical ultrasound images is more challenging. In this paper, the Two-Stage GAN network model was adjusted by CycleGAN generation and unsupervised learning methods, and the Two-Stage ZSSR (“Zero-Shot” Super-Resolution) CycleGAN network was proposed. The objective evaluation indexes PSNR and SSIM were raised to 40.8079 and 0.9953. The visual effect was also significantly improved.

Additional file 1: Table S1. Group: 1 represents benign ovarian tumor, 2 represents BOT. M: 0 represents premenopausal status, 1 represents postmenopausal status. U represents ultrasound score.

Automatic tumor segmentation of breast ultrasound (BUS) image is quite challenging due to the complicated anatomic structure of breast and poor image quality. Most tumor segmentation approaches achieve good performance on BUS images collected in controlled settings; however, the performance degrades greatly with BUS images from different sources. Tumor saliency estimation (TSE) has attracted increasing attention to solving the problem by modeling radiologists' attention mechanism. In this paper, we propose a novel hybrid framework for TSE, which integrates both high-level domain-knowledge and robust low-level saliency assumptions and can overcome drawbacks caused by direct mapping in traditional TSE approaches. The new framework integrated the Neutro-Connectedness (NC) map, the adaptive-center, the correlation and the layer structure-based weighted map. The experimental results demonstrate that the proposed approach outperforms state-of-the-art TSE methods., Comment: 6 pages and 8 figures, Conference paper and accepted by ICPR 2018

The purpose of this study was to assess the clinical value of ultrasound (US) features of breast lesions for predicting the risk of axillary lymph node metastasis in patients with breast cancer.In this retrospective study, 425 patients with breast cancer were recruited, and their preoperative US features and postoperative pathologic results were collected. The association of these US features of breast cancer with axillary lymph node metastasis was determined by univariate and multivariate analyses.Among the 425 patients, 200 (47.1%) had axillary lymph node metastasis, and 225 (52.9%) did not. The parameters of tumor shape, color Doppler flow imaging grades, histologic grade, and E-cadherin level were significantly and independently associated with axillary lymph node metastasis (P .05 for all).Axillary lymph node metastasis was prone to happen in patients with US features of an irregular tumor shape and higher color Doppler flow imaging grades. Ultrasound imaging provides a promising tool for predicting axillary lymph node metastasis in patients with breast cancer.

The purpose of our study was to assess the potential clinical value of ultrasound imaging in predicting risk category in patients with breast cancer. Three hundred thirty-six patients were enrolled and divided into a high-risk group (99, 29.5%) and mid- to low-risk group (237, 70.5%) according to the St. Gallen risk criteria. All data were retrospectively collected to analyze correlations between ultrasound features and risk category. The results revealed that the ultrasound features of irregular shape (p = 0.002), vertical growth orientation (p = 0.002), angular contour (p = 0.022) and high color Doppler flow imaging grade (p = 0.001) tended to be present in images of the high-risk group. Therefore, tumor ultrasound features should be recognized as an ideal option for determination of risk category in patients with breast cancer.

Aim To summarize the ultrasonic features of isolated fallopian tube torsion (IFTT) by retrospectively analyzing cases presenting at our hospital. Methods This analysis was approved by the ethical committee of our hospital. Medical records of surgically proven IFTT patients admitted to our hospital since 2002 were collected. Clinical features and preoperative diagnoses of the patients were analyzed retrospectively. Ultrasonic images were reviewed and characteristics, including location, size, shape, echo and vascularity of the fallopian tube, were summarized. Results Eleven patients with IFTT were studied. No obvious association was found between the torsion and menstruation cycle. Only four accurate diagnoses were made before the operation. Degrees of torsion ranged 360–2160°. Most of them (9/11, 81.8%) were greater than 720°. Sonograms of the 11 patients could be classified into four types: cystic masses, tube-like structures, heterogeneous masses and whirlpool signs. Cystic masses were the most commonly seen type (4/11, 36.4%), followed by tubular structures (3/11, 27.3%). Whirlpool sign was believed to be the most specific sign in diagnosing IFTT. Conclusion Through review of the authors' experiences, it is possible to diagnose IFTT preoperatively by ultrasound. Sonograms of the IFTT could be divided into four types while clinical significance of this classification requires further confirmation.

UNLABELLED PURPSOE: To investigate the diagnostic value of quantitative analysis of a tissue diffusion and virtual touch tissue imaging quantification (VTIQ) technique with acoustic radiation force impulse (ARFI) elastography for assessing enlarged cervical lymph nodes. MATERIALS AND METHODS Fifty-six enlarged cervical lymph nodes confirmed by pathologic diagnoses were covered in the study. According to the results of pathologic diagnosis, patients were classified into benign and malignant groups. All the patients were examined by both conventional ultrasonography and elastography. AREA% and shear wave velocity (SWV) in ROI of different groups were calculated and compared using ROC curves. Cut-off points of AREA% and SWV were determined with receiver operating characteristic curves. RESULTS Final histopathological results revealed 21 cases of benign and 35 cases of malignant lymph nodes. The mean values of AREA% and SWV in benign and malignant groups were 45.0 ± 17.9% and 2.32 ± 0.57 m/s, and 61.3 ± 21.29% and 4.36 ± 1.25)m/s, respectively. For the parameters of elastography, "AREA%"and SWV demonstrated significant differences between groups (p=0.002). AREA% was positively correlated with SWV with a correlation coefficient of 0.809 (P

Objectives The purpose of this study was to assess the clinical value of ultrasound (US) features of breast lesions for predicting the risk of axillary lymph node metastasis in patients with breast cancer. Methods In this retrospective study, 425 patients with breast cancer were recruited, and their preoperative US features and postoperative pathologic results were collected. The association of these US features of breast cancer with axillary lymph node metastasis was determined by univariate and multivariate analyses. Results Among the 425 patients, 200 (47.1%) had axillary lymph node metastasis, and 225 (52.9%) did not. The parameters of tumor shape, color Doppler flow imaging grades, histologic grade, and E-cadherin level were significantly and independently associated with axillary lymph node metastasis (P

The objective of this study was to investigate the feasibility of using 3-D transvaginal tomographic ultrasound imaging (TUI) to stage patients with cervical carcinoma. Eighty women with cervical cancer who underwent transvaginal TUI examinations were enrolled. In all patients, cancer was confirmed pre-operatively by pathologic examination. Staging on the basis of clinical features, ultrasonography and magnetic resonance imaging was performed according to the International Federation of Gynecology and Obstetrics (FIGO) staging system. Clinical, TUI and magnetic resonance imaging staging was compared with that based on histology. Depth of invasion into the stroma was measured by TUI in 52 cases and compared with pathologic results. An interclass correlation coefficient was used to analyze reproducibility. In total, all 80 patients underwent surgical treatment. The accuracy of pre-operative staging, compared with histologic findings, was 92.50% for TUI, 82.50% for magnetic resonance imaging and 78.75% for clinical examination. The mean depth of lesions as measured with TUI was 12.5 ± 6.2 mm (range: 3.5–40.0 mm), and that measured on histology was 10.5 ± 8.0 mm (range: 3.0–40.0 mm). The interclass correlation coefficient of the two methods was 0.933 (95% confidence interval: 0.887–0.961). Pre-operative TUI is promising as a method for pre-operative staging of cervical carcinomas. TUI can also reliably assess lesion depth.

Bone marrow mesenchymal stem cells (BMSCs) emerge as a promising approach for treating heart diseases. However, the effects of BMSCs-based therapy on cardiac electrophysiology disorders after myocardial infarction were largely unclear. This study was aimed to investigate whether BMSCs transplantation prevents cardiac arrhythmias and reverses potassium channels remodelling in post-infarcted hearts. Myocardial infarction was established in male SD rats, and BMSCs were then intramyocardially transplanted into the infarcted hearts after 3 days. Cardiac electrophysiological properties in the border zone were evaluated by western blotting and whole-cell patch clamp technique after 2 weeks. We found that BMSCs transplantation ameliorated the increased heart weight index and the impaired LV function. The survival of infarcted rats was also improved after BMSCs transplantation. Importantly, electrical stimulation-induced arrhythmias were less observed in BMSCs-transplanted infarcted rats compared with rats without BMSCs treatment. Furthermore, BMSCs transplantation effectively inhibited the prolongation of action potential duration and the reduction of transient and sustained outward potassium currents in ventricular myocytes in post-infarcted rats. Consistently, BMSCs-transplanted infarcted hearts exhibited the increased expression of K(V)4.2, K(V)4.3, K(V)1.5 and K(V)2.1 proteins when compared to infarcted hearts. Moreover, intracellular free calcium level, calcineurin and nuclear NFATc3 protein expression were shown to be increased in infarcted hearts, which was inhibited by BMSCs transplantation. Collectively, BMSCs transplantation prevented ventricular arrhythmias by reversing cardiac potassium channels remodelling in post-infarcted hearts.

Cerebral atherosclerosis vascular stenosis is a common etiology for ischemic stroke and a major factor in recurrent stroke and vascular mortality. Recent studies suggest that renalase plays a role in hypertension and ischemic stroke, and may be involved in atherosclerosis. The aim of the present study was to investigate whether there were correlations between single-nucleotide polymorphisms (SNPs) in the renalase gene and severity of intracranial cerebral atherosclerotic vascular stenosis in ischemic stroke patients as determined by imaging. A total of 212 ischemic stroke patients and 244 healthy controls from the north Chinese Han population were enrolled in this study. Polymerase chain reaction and ligase detection reaction were used for SNP analysis. We classified the case samples by severity of the intracranial cerebral atherosclerotic vascular stenosis. Allele, genotype, and haplotype were analyzed in cases and controls, and logistic regression was used to adjust for bias due to conventional stroke risk factors. The allele and the genotype of rs10887800 in the renalase gene were both associated with severe intracranial cerebral atherosclerotic vascular stenosis (p = 0.013 and p = 0.049, respectively). No association was observed with severity for SNP rs2576178 or SNP rs2296545. Our findings show that the SNP rs10887800 in the renalase gene is closely associated with severe intracranial cerebral atherosclerotic vascular stenosis in ischemic stroke patients of north Chinese Han origin.

Ischemic stroke is a leading cause of death and adult disability worldwide. Recent research suggests that renalase is strongly associated with heart disease and hypertension and may play a role in ischemic stroke. In this study, we investigate the genetic association between renalase and ischemic stroke in a northern Chinese Han population. We genotyped single-nucleotide polymorphisms of the renalase gene in 507 ischemic stroke patients and 503 gender-matched controls from a northern Chinese Han population, and we classified these case samples in depth by complications with diabetes or hypertension and by ischemic stroke subtypes. We undertook allele, genotype, and haplotype association studies in all the cases and in the subgroups, as well as multiple factor analysis by logistic regression. rs10887800 and rs2576178 were significantly associated with ischemic stroke with hypertension by logistic regression (p = 0.041 and p = 0.038, respectively). The recessive model showed a strong association of rs2296545 with ischemic stroke patients in hypertension subgroups (OR = 1.927, 95 % CI = 1.012–3.669, p = 0.046). The renalase gene is closely related to ischemic stroke in this northern Chinese Han population, suggesting that renalase may be mechanistically involved in stroke pathology.

Background Distant metastases stemming from a papillary thyroid carcinoma (PTC) are quite rare. Here we report an exceptional case of PTC presenting with cervical lymphatic and uterine metastases. This is the first case report of a PTC with uterine involvement. Case presentation A 60-year-old Chinese woman came to our hospital complaining of discomfort in the throat that she had been experiencing for about half a month. PTC and cervical lymphatic metastasis were diagnosed after ultrasound examinations. A massive heterogeneous mass was found beside the uterus during the pre-operative checkup and a diagnosis of ovarian carcinoma was suspected after a thorough case discussion. However, it proved to be a metastasis from the PTC as determined by pathological and immunohistochemical examinations after the operation. The patient declined further treatments. She was followed for 22 months with no sign of recurrence detected. Conclusions In this report, an unusual case of PTC was presented. The patient had not only regional lymphatic metastasis, but also had a massive metastasis in the uterine corpus, which was initially misdiagnosed as ovarian carcinoma. This case is of interest because of its rarity and exceptionally good prognosis. The reason for the misdiagnosis was attributed to overlooking the possibility of a distant metastasis coming from a PTC. This case raises the issue that completing an iodine-131 scan before operating on patients with PTC may be warranted.