Search

Purpose: Bone metastasis (BM) lesions are always considered to be non-measurable. In this study, we established a method that proved to be measurable the subtypes of BM by in vivo advanced imaging including bone scintigraphy (BS) and single-photon emission computed tomography (SPECT) with 99mTechnetium-methylene diphosphonate (99mTc-MDP) fused computed tomography (CT), combined with skeletal radiography (XR). Methods: One retrospective clinical audit was investigated on lung cancer patients who suffer from BM. The audit includes early static planar, later Whole-body skeleton images and a random sample of 198 patients who were performed on fusion of SPECT/CT scanners. Among them, 108 patients with BM were definite histologic diagnosed with lung cancer and 257 bone lesions were classified. A preclinical precision study developed a procedure of hot spot-based bone imaging screening experimental BM clone by performing anesthesia, intracardiac injection of five human lung cancer cells on the immunodeficient mice, reanimate, in vivo imaging and in vitro experiments (cell culture, histology, karyotype analysis, microarrays, real-time polymerase chain reaction assay (RT-PCR), and immunohistochemistry (IHC)). An osteolytic metastatic clone MDA-MB-231Bo (231Bo) was compared. Irradiation damage was tested. Results: The clinical reports showed the incidence was primarily osteoblastic lesions, followed by mixed lesions and osteolytic. This study does not only confirmed the in vivo advanced imaging could be used to monitor and measure the subtypes of BM from clinical and preclinical data but also have ideal images (including movies), unique cell lines same as homo-sapiens, two markers associated with BM, TMEM100 and CDH1, as well as data from five pairs (BM clones and their parental cells) of lung cancer cells that released into GEO for further research. Ionizing radiation is controlled at its best. Conclusion: The studies have demonstrated that bone lesions of 1mm or larger can be detected in mouse models. Osteolytic lesions were revealed that a flow of metastatic cells out of the destroying cortical bone gap to form a soft tissue tumor. This study and GEO data are applicable to the development of relevant research and treatments.

The biomass-based carbon materials with heteroatomic regulation, hierarchical structure, remarkable porosity, excellent conductivity and good stability are promising energy materials for supercapacitors. Herein, we designed and synthesized the multidimensional morphology controllable carbon materials by activation approach using pervasive watermelon rind as the carbon source. Interestingly, the carbon materials revealed diverse dimensional morphology characterizations for 3D hierarchical honeycomb nanoporous carbon architecture (CNP), 2D carbon nanoflake model (CNF) and 1D needle-shaped carbon nanobars (CNB) with/without different dopants. The CNB energy material indicated a large specific area for 879.18 m2 g−1, and the flexible electrode showed excellent areal capacitance for ∼2245 F cm−2 at 1 mA cm−2. Moreover, the constructed flexible symmetric device shows outstanding cycling stability for 99.5% after 20,000 cycles, superior energy density for 0.12 mWh cm−2 at a power density of 0.52 mW cm−2. This work provides a facile and sustainable approach to synthesize high-performance energy materials for supercapacitor.

Real-time hand gesture recognition in complex environments has many challenges, such as poor real-time performances and robustness to environmental changes. This paper takes the hand gesture control of the unmanned vehicle as the application background, and focuses on the gesture detection and recognition of video streams based on deep learning in the complex environment. In this paper, we detect the hand in a complex environment by training the ssd_mobilenet model, and initialize the tracking with kalman filter. Then, we detect the hand keypoints by following the architecture of Convolutional Pose Machines (CPMs), in order to obtain the belief maps for all keypoints that are used as the train sets of Convolutional Neural Networks (CNNs). Finally, based on the results obtained by our classification, this paper proposes a method of multi-frame recursion to minimize the influences of redundant frames and error frames. In this paper, eight kinds of gestures for controlling vehicle are identified. The experimental results show that our method can successfully realize real-time hand gesture recognition in the video streams. The recognition accuracy can reach 96.7%, and the average recognition speed reaches 12 fps, which basically meets the real-time requirements and successfully applys to mobile terminals such as TX2 for engineering practice.

Monocular visual odometry as the front end of SLAM, its task is to estimate the movement of cameras between adjacent images. One of the most important steps in visual odometry is to find feature correspondence between neighboring frames. In finding the feature matches, the search region can not be too small in order to ensure a large number of matches to be found. On the other hand, it could not be too large, otherwise it would generate many false matches and will deteriorate the entire system performance. In this paper, we propose to add a dense optical flow computation step before the matching step. The dense optical flow is computed through optimization from the whole image. The image context will help to ensure that its result will not deviate too far away from the true position, and thus provides an initial position estimate for the feature matches. Experiments on KITTI dataset show that the proposed approach can indeed narrow down the search range for finding feature correspondence while still being able to find a large number of correct matches.

Multi-sensor image registration can combine information obtained by multimodality sensors and describe the scene more accurately, thus making it a hot research topic in the field of image processing. In this paper, we propose a novel approach for highly accurate multi-sensor image registration based on undirected graph model. In the proposed approach, the source image is divided into five patches and each patch slides in the target image to obtain a correlation surface which describes the similarity between the patch and the corresponding region of the target image. To promote the global consistency of the registration result, we treat every point on the correlation surface as a state of a node of UGM, while the similarity of the motion fields in the overlapping area of two adjacent patches is viewed as an edge. Through UGM, we can get the best combination of neighboring patches, which simultaneously ensure the accuracy of each patch's motion field and the smooth transition between neighboring patches.

In this paper, we propose a new approach to refine the estimated transmission map which based on dark channel prior. Traditionally, the raw transmission can be refined by guided filter using hazy image, but atomizing blurred salient edges and made transmission map including lots of redundant texture details which represented error information of depth. We obtain transmission map under the small and large scale mask using the minimum filtering. In order to have abundant high frequent information as well as accurate low frequent information from transmission. Firstly a new edge-preserving smoothing method named L0 smoothing will be conducted to diminish insignificant details in the transmission map under small scale mask, L0 smoothing can eliminate a manageable degree of low-amplitude structures and globally control the number of non-zero gradients. Then, we use guided filter based on a local linear model for information fusion of different frequent information. We demonstrate that the proposed algorithm can preserve prominent edge and keep the transmission map precise in a more appropriate area.

Unstructured road detection is a key step of the Unmanned Guided Vehicle (UGV) system for road following. However, current vision-based unstructured road detection algorithms are usually affected by continuously changing backgrounds, different road types (shape, color), variable lighting conditions and weather conditions. Therefore, a road distribution model is theoretically analyzed and experimentally generated to help detecting unstructured roads. Global map and global positioning system (GPS) information are used to choose the corresponding road distribution model. Two traditional algorithms, support vector machine (SVM) and k-nearest neighbor (KNN), are carried out to verify the helpfulness of the proposed algorithm. The proposed algorithm has been evaluated by different types of unstructured roads and the experimental results show its effectiveness.

Unstructured road detection is a key step of the Unmanned Guided Vehicle (UGV) system for road following. In traditional methods, kinds of features are combined to improve the effect, but few of them explains the reason to choose feature descriptors. In this paper, we try to find a way that how to choose feature descriptors. First, support vector machine technology is used to analyze the importance among these common feature descriptors during the road detection process. A series of features are analyzed to show their ability to detect road surface from background. Then, a mass of experiment results verify the correct of our analysis. At last, a novel unstructured road detection algorithm based on hybrid features is proposed. Experiments conducted on actual roads illustrate the effective performance of the proposed algorithm. information Unstructured road detection is a key step of the Unmanned Guided Vehicle (UGV) system for road following. However, current vision-based unstructured road detection algorithms are usually affected by continuously changing backgrounds, different road types (shape, color), variable lighting conditions and weather conditions. Therefore, a confidence map of road distribution, one of contextual information cues, is theoretically analyzed and experimentally generated to help the detection of unstructured road. The RGB feature + SVM algorithm and the RGB feature + KNN algorithm are carried out to verify the helpful of the proposed contextual cue. The proposed confidence map has been evaluated by different type of unstructured roads and the experimental results show its effectiveness.

The recurrence, metastasis and multidrug resistance (MDR) in lung cancer are the tough problems worldwide. This study was to establish a novel chinese lung adenocarcinoma cell line with high metastasis potential for exploring the mechanism of reccurrence, development and MDR in lung cancer.The cell came from the abdominal dropsy of a fifty-six years old female patient with lung adenocarcinoma and the tumor markers CA125, CYFRA21-1, CEA, NSE were detected to be higher secretion by radioimmunoassay in the abdominal dropsy. Tumorigenicity of immunodeficient mice was confirmed in 8th passage. The cell growth curve was mapped. Analysis of chromosome karyotype was tested. The gene expression was measured by real-time quantitative PCR.The tumorigenesis rate started at 8th passage in 3/10 immunodeficient mice via subcutaneously and the fully tumorigenicity was at 11th passage as well as later passages. Under the microscope, the cell showed oval-shap and adherence. The chromosome karyotype analysis of the cells was sub-triploid. Approximately 1*10(6) and 1.5*10(6) cancerous cells were injected into left cardiac ventricle and tail vein of immunodeficient mice respectively. The results showed multiorgan metastasis in the mice after three-four weeks, including mandible, scapula, humerus, vertebral column, femur, rib and brain, liver, adrenal gland, pulmonary in the mice after inoculation. The bone metastasis rate was 100% in the tumor bearing mice by bone scintigraphy and pathology. Quantitative real-time PCR was used to examined and compared with SPC-A-1 lung adenocarcinoma, ESM1, VEGF-C, IL-6, IL-8, AR genes were overexpressed. The novel cell was named CPA-Yang2The characteristics of novel strain CPA-Yang2 is a highly metastasis cell line of Chinese lung adenocarcinoma. It has stable traits, highly metastasis ability and maybe is a MDR lung cancerous cell line. Of course, it's a good experimental model for lung cancer research.

Background Bone metastasis is one of the most common clinical phenomena of late stage lung cancer. A major impediment to understanding the pathogenesis of bone metastasis has been the lack of an appropriate animal and cell model. This study aims to establish human lung adenocarcinoma cell line with highly bone metastases potency with 99m Tc-MDP bone scintigraphy. Methods The human lung adenocarcinoma cancer cells SPC-A-1 were injected into the left cardiac ventricle of NIH-Beige-Nude-XID (NIH-BNX) immunodeficient mice. The metastatic lesions of tumor-bearing mice were imaged with 99m Tc-MDP bone scintigraphy on a Siemens multi-single photon emission computed tomography. Pinhole images were acquired on a GZ-B conventional gamma camera with a self-designed pinhole collimator. The mice with bone metastasis were sacrificed under deep anesthesia, and the lesions were resected. Bone metastatic cancer cells in the resected lesions were subjected for culture and then reinoculated into the NIH-BNX mice through left cardiac ventricle. The process was repeated for eight cycles to obtain a novel cell subline SPC-A-1BM. Real-time polymerase chain reaction (PCR) was used to compare the gene expression differences in the parental and SPC-A-1BM cells. Results The bone metastasis sites were successfully revealed by bone scintigraphy. The established bone metastasis cell line SPC-A-1BM had a high potential to metastasize in bone, including mandible, humerus, thoracic vertebra, lumbar, femur, patella, ilium and cartilage rib. The expression level of vascular endothelial growth factor gene family, Bcl-2 and cell adhesion-related genes ECM1 , ESM1 , AF1Q , SERPINE2 and FN1 were examined. Gene expression difference was found between parental and bone-seeking metastasis cell SPC-A-1BM, which indicates SPC-A-1BM has metastatic capacity vs. its parental cells. Conclusion SPC-A-1BM is a bone-seeking metastasis human lung adenocarcinoma cell line. Bone scintigraphy may be used as an accurate, sensitive, noninvasive tool to detect experimental bone metastases in intact live NIH-BNX mice.

Virtual reality is an effective way for visualization of science experiments. This paper presents a VR enhanced display system for the experiment of HgCdTe(MCT) single crystal growth by Bridgman method in two steps: In the first step, we construct a numerical simulation model (time dependent) based on the heat transfer in furnace and ampoule. This gives information about the most important parameter during the whole growth process: the temperature distribution in furnace and ampoule. Then, with the obtained temperatures, we have developed a VR environment in the second step to enhance the display of the key parameters information of crystal growth. This helps the researcher to understand the process of crystal growth intuitively.