Fully automatic tumor segmentation of breast ultrasound images with deep learning.

BACKGROUND: Breast ultrasound (BUS) imaging is one of the most prevalent approaches for the detection of breast cancers. Tumor segmentation of BUS images can facilitate doctors in localizing tumors and is a necessary step for computer-aided diagnosis systems. While the majority of clinical BUS scans are normal ones without tumors, segmentation approaches such as U-Net often predict mass regions for these images. Such false-positive problem becomes serious if a fully automatic artificial intelligence system is used for routine screening. METHODS: In this study, we proposed a novel model which is more suitable for routine BUS screening. The model contains a classification branch that determines whether the image is normal or with tumors, and a segmentation branch that outlines tumors. Two branches share the same encoder network. We also built a new dataset that contains 1600 BUS images from 625 patients for training and a testing dataset with 130 images from 120 patients for testing. The dataset is the largest one with pixel-wise masks manually segmented by experienced radiologists. Our code is available at https://github.com/szhangNJU/BUS_segmentation. RESULTS: The area under the receiver operating characteristic curve (AUC) for classifying images into normal/abnormal categories was 0.991. The dice similarity coefficient (DSC) for segmentation of mass regions was 0.898, better than the state-of-the-art models. Testing on an external dataset gave a similar performance, demonstrating a good transferability of our model. Moreover, we simulated the use of the model in actual clinic practice by processing videos recorded during BUS scans; the model gave very low false-positive predictions on normal images without sacrificing sensitivities for images with tumors. CONCLUSIONS: Our model achieved better segmentation performance than the state-of-the-art models and showed a good transferability on an external test set. The proposed deep learning architecture holds potential for use in fully automatic BUS health screening.

A deformable model for navigated laparoscopic gastrectomy based on finite elemental method.

Background: Accurate registration for surgical navigation of laparoscopic surgery is highly challenging due to vessel deformation. Here, we describe the design of a deformable model with improved matching accuracy by applying the finite element method (FEM).Material and methods: ANSYS software was used to simulate an FEM model of the vessel after pull-up based on laparoscopic gastrectomy requirements. The central line of the FEM model and the central line of the ground truth were drawn and compared. Based on the material and parameters determined from the animal experiment, a perigastric vessel FEM model of a gastric cancer patient was created, and its accuracy in a laparoscopic gastrectomy surgical scene was evaluated.Results: In the animal experiment, the FEM model created with Ogden foam material exhibited better results. The average distance between the two central lines was 6.5mm, and the average distance between their closest points was 3.8 mm. In the laparoscopic gastrectomy surgical scene, the FEM model and the true artery deformation demonstrated good coincidence.Conclusion: In this study, a deformable vessel model based on FEM was constructed using preoperative CT images to improve matching accuracy and to supply a reference for further research on deformation matching to facilitate laparoscopic gastrectomy navigation.

The lncRNA TDRG1 promotes cell proliferation, migration and invasion by targeting miR-326 to regulate MAPK1 expression in cervical cancer.

BACKGROUND: Recently, lncRNA-Testis developmental related gene 1 (TDRG1) was proved to be a key modulator in reproductive organ-related cancers. The biological role of TDRG1 in cervical cancer (CC) progression remains largely unknown. METHOD: Real-time PCR (qRT-PCR) examined the expression level of TDRG1, microRNA (miR)-326 and MAPK1 mRNA. OS tissues and corresponding relative normal tissues, as well as CC cell lines and normal cell line Ect1/E6E7 were collected to determine the expression of TDRG1 in CC. MTT, colony formation, wound-healing, transwell and flow cytometer assay detected the influence of TDRG1 and miR-326 on CC cells growth, metastasis and apoptosis. Western blot examined proteins level. Bioinformatics, RNA pull-down assay, RNA immunoprecipitation and dual-luciferase reporter assays detected the molecular mechanism of TDRG1 in CC. Xenograft tumour model was established to determine the role of TDRG1 in vivo. RESULTS: The expression of TDRG1 was significantly increased in CC tissues and cell lines compared with normal tissue and normal cell line respectively and its expression was associated with clinicopathological characteristics of CC patients. Knockdown of TDRG1 inhibited the cell proliferation, migration and invasion in Hela and SIHA cells. Moreover, TDRG1 directly interacted with miR-326, and the inhibition effect on cell growth and metastasis induced by TDRG1 siRNA can be abrogated by miR-326 silencing by its inhibitor in Hela and SIHA cells. Further, MAPK1 was proved to be a direct target of miR-326, and its expression was negatively regulated by miR-326 while positively modulated by TDRG1. CONCLUSION: TDRG1 acts as a competing endogenous lncRNA (ceRNA) to modulate MAPK1 by sponging miR-326 in CC, shedding new light on TDRG1-directed diagnostics and therapeutics in CC.

TRUSS inhibition protects against high fat diet (HFD)-stimulated brain injury by alleviation of inflammatory response.

High fat diet (HFD)-induced obesity is associated with insulin resistance (IR) and other chronic, diet associated illnesses, including neuroinflammation and brain injury. However, the involvement of inflammatory response in HFD-elicited central nerve injury has yet to be fully determined. Recent studies have indicated that tumor necrosis factor receptor-associated ubiquitous scaffolding and signaling protein (TRUSS), also known as TRPC4AP, plays an essential role in regulating inflammation via the meditation of NF-κB signaling. In the present study, we attempted to explore the effects of TRUSS on HFD-induced brain injury in the wild type mice (TRUSS+/+) or TRUSS-knockout mice (TRUSS-/-). The results suggested that TRUSS deletion attenuated HFD-induced cognitive impairments in mice. HFD-elicited metabolic disorders were also highly improved by the loss of TRUSS, as evidenced by the reduced serum glucose and insulin levels, as well as the lipid deposition in liver tissues. In addition, HFD-triggered brain injury was markedly alleviated by the TRUSS ablation, as proved by the reduction of GFAP and Iba1 expressions in hippocampus and hypothalamus. Moreover, TRUSS-/- mice exhibited a significant decrease in the expression of pro-inflammatory cytokines, accompanied with the inactivation of IKKα/IκBα/NF-κB pathway. At the same time, HFD-induced dyslipidemia was also alleviated by the loss of TRUSS. The in vitro study verified the protective effects of TRUSS-suppression against HFD-induced central nerve injury and hepatic steatosis by restraining the inflammatory response. In summary, our data indicated that TRUSS participated in metabolic syndrome-induced brain injury and pointed to the repression of TRUSS as a promising strategy for cognitive deficits therapy.

Preparation and physicochemical properties of the complex of naringenin with hydroxypropyl-beta-cyclodextrin.

In this study a complex of naringenin with hydroxypropyl-beta-cyclodextrin (HP-beta-CD) was prepared to improve the hydrophilicity of naringenin. The physicochemical properties of the complex were analyzed by ultraviolet-visible spectrometry (UV), infrared spectrometry (IR), X-ray diffractometry (XRD), differential scanning calorimetry (DSC). The result showed that naringenin had been molecularly dispersed in the HP-beta-CD matrix, not forming a new compound and HPLC analysis showed that the solubility of naringenin in water was enhanced from 4.38 microg/mL to 1,272.31 microg/mL.