Search

Spatial target pose estimation with deep learning has garnered increasing attention in recent years. However, the existing methods in this field suffer from poor generalization. In this study, we propose a robust and reliable pose estimation method for spatial targets. The method aims to achieve keypoints filtrating. It involves a detection network tasked with identifying the target area, while the subsequent stage employs a classification network to regress keypoints from the detected target area. To improve the accuracy of pose estimation, we leverage spatial target geometric constraints to formulate 2-D–3-D keypoints equations for an initial pose. Then, we create a nonlinear optimization equation based on the confidence of 2-D keypoints and accomplish nonlinear refinement. We conduct extensive experiments on commonly used datasets and demonstrate the effectiveness of the proposed method. Furthermore, thanks to the effectiveness of keypoints filtrating and nonlinear refinement, the proposed method is robust with challenging scenarios and domain bias.

In the testing of chips, defect diagnostics in X-ray images of packaging chips is mainly performed by humans, which is time-consuming and inefficient. To overcome the abovementioned problems, a novel intelligent defect diagnostics system based on hybrid deep learning for chip X-ray images was proposed. The system consists of four successive stages: image segmentation and normalization, image reconstruction and defect detection, contour matching, and qualification diagnosis. The first stage is used to localize the external contours of the target chip and remove extraneous backgrounds through the improved UNet. Then, considering the variety of defects and the complexity of labeling, an unsupervised learning model is designed to reconstruct defect-free images to detect defects, which requires only normal samples for training. Third, the multicomponent template matching based on structural prior is used to localize the internal contours of the chip. In the final stage, the qualification is diagnosed based on the previous results through the Floyd–Warshall algorithm. The effectiveness and robustness of the proposed methods are verified by experiments on real-world inspection lines. The experimental results demonstrate that the developed system can successfully perform fault diagnostics tasks, achieving a judgment accuracy of 92.5%.

ABSTRACTAngiopoietin-like 3 (ANGPTL3) has been uncovered to play an oncogenic role in several kinds of human malignancies. Nevertheless, whether ANGPTL3 functions in cervical cancer (CC) has not yet been reported. This paper is intended to explore the impact of ANGPTL3 on CC cells and elucidate the potential mechanism. In this study, quantitative real-time polymerase chain reaction (qRT-PCR) and Western blot were performed to analyze the ANGPTL3 expression. Western blot was also performed to examine integrin αvβ3 protein level. Cell proliferation was evaluated by MTT assay, EdU staining and Western blot analysis. In addition, the migratory and invasive abilities of cells were, respectively, estimated by wound healing and transwell assays. Tube formation assay was performed to determine endothelial cell angiogenesis. Levels of vascular endothelial growth factor (VEGF) and vascular endothelial growth factor receptor 2 (VEGFR2) were measured by ELISA. As a result, ANGPTL3 expression was significantly higher in CC cells relative to that in normal cervical cells. Silencing of ANGPTL3 suppressed cell proliferation, migration and invasion. Besides, downregulation of ANGPTL3 inhibited human umbilical vein endothelial cell (HUVEC) angiogenesis and repressed protein level of integrin alpha v beta 3 (αvβ3). Upregulation of αvβ3 offsets the inhibitory effect of ANGPTL3 on proliferation, migration and invasion in CC cells. Upregulated expression of αvβ3 promoted blood vessel formation and secretions of VEGF and VEGFR2. In conclusion, ANGPTL3 silencing may serve as a tumor suppressor in CC through integrin αvβ3, which provides a potentially novel therapeutic target for patients with CC.

Integrin αvβ6 is expressed at an undetectable level in normal tissues, but is remarkably upregulated during many pathological processes, especially in cancer and fibrosis. Noninvasive imaging of integrin αvβ6 expression using a radiotracer with favorable in vivopharmacokinetics would facilitate disease diagnosis and therapy monitoring. Through disulfide-cyclized method, we synthesized in this study, a new integrin αvβ6-targeted cyclic peptide (denoted as cHK), and radiolabeled it with 99mTc. The ability of the resulting radiotracer 99mTc–HYNIC–cHK to detect integrin αvβ6 expression in pancreatic cancer xenografts and idiopathic pulmonary fibrosis was evaluated using small-animal single-photon emission computed tomography (SPECT)/computed tomography (CT). 99mTc–HYNIC–cHK showed significantly improved in vivometabolic stability compared to the linear peptide-based radiotracer 99mTc–HYNIC–HK. 99mTc–HYNIC–cHK exhibited similar biodistribution properties to 99mTc–HYNIC–HK, but the tumor-to-muscle ratio was significantly increased (2.99 ± 0.87 vs. 1.82 ± 0.27, P < 0.05). High-contrast images of integrin αvβ6-positive tumors and bleomycin-induced fibrotic lungs were obtained by SPECT/CT imaging using 99mTc–HYNIC–cHK. Overall, our studies demonstrate that 99mTc–HYNIC–cHK is a promising SPECT radiotracer for the noninvasive imaging of integrin αvβ6 in living subjects.

Effective cancer therapy depends not only on destroying the primary tumor but also on conditioning the host immune system to recognize and eliminate residual tumor cells and prevent metastasis. In this study, a tumor integrin αvβ6-targeting peptide (the HK peptide)-functionalized graphene oxide (GO) was coated with a photosensitizer (HPPH). The resulting GO conjugate, GO(HPPH)-PEG-HK, was investigated whether it could destroy primary tumors and boost host antitumor immunity. We found that GO(HPPH)-PEG-HK exhibited significantly higher tumor uptake than GO(HPPH)-PEG and HPPH. Photodynamic therapy (PDT) using GO(HPPH)-PEG suppressed tumor growth in both subcutaneous and lung metastatic mouse models. Necrotic tumor cells caused by GO(HPPH)-PEG-HK PDT activated dendritic cells and significantly prevented tumor growth and lung metastasis by increasing the infiltration of cytotoxic CD8+T lymphocytes within tumors as evidenced by in vivooptical and single-photon emission computed tomography (SPECT)/CT imaging. These results demonstrate that tumor-targeted PDT using GO(HPPH)-PEG-HK could effectively ablate primary tumors and destroy residual tumor cells, thereby preventing distant metastasis by activating host antitumor immunity and suppressing tumor relapse by stimulation of immunological memory.

There is growing interest in the development of radiolabeled multivalent ligands because of their higher tumor uptake versus that of the corresponding monomer. This report presents the synthesis of a [Tyr3]octreotide dimer conjugate, HYNIC-E([Tyr3]octreotide)2{HYNIC-TOC2, HYNIC = 6-[2-(2-sulfonatobenzaldehyde)hydrazono]nicotinyl}, and its biological evaluation in the AR42J tumor model. The binding affinity of HYNIC-TOC2for somatostatin receptor subtype 2 (SSTR2) was determined in AR42J rat pancreatic cancer cells, using 125I-[Tyr3]octreotide as the radiotracer. 99mTc-HYNIC-TOC2was prepared by using tricine and EDDA as coligands (EDDA = ethylenediamine-N,N′-diacetic acid). Biodistribution and γ imaging were performed in nude mice bearing AR42J tumors. 99mTc-HYNIC-TOC2was obtained in >95% labeling yield with favorable stability. Compared with those of HYNIC-TOC (IC50= 3.74 ± 0.82 nM), HYNIC-TOC2showed significantly increased SSTR2 binding affinity (IC50= 0.74 ± 0.19 nM), and 99mTc-HYNIC-TOC2showed significantly increased tumor uptake [13.31 ± 3.14%ID/g vs 5.32 ± 0.94%ID/g 1 h postinjection (p.i.) and 12.05 ± 2.92%ID/g vs 5.87 ± 1.96%ID/g 4 h p.i.]. Although the level of accumulation of 99mTc-HYNIC-TOC2in kidneys was significantly increased (94.40 ± 6.51%ID/g vs 32.27 ± 4.51%ID/g 1 h p.i.), this high uptake was inhibited by the injection of l-lysine before the administration of 99mTc-HYNIC-TOC2(30.99 ± 5.05%ID/g 1 h p.i.) while tumor uptake decreased only slightly. Consistent with biodistribution data, in vivoplanar γ imaging showed that the tumors were clearly visualized, while the background signal was much weaker except for that of the kidneys and bladder. The new radiotracer 99mTc-HYNIC-TOC2with a higher binding affinity and good stability was designed and evaluated. The higher tumor uptake of 99mTc-HYNIC-TOC2suggests that 90Y/177Lu-labeled TOC2might have an advantage for the radiotherapy of SSTR2-positive tumors. These data merit the translation of 99mTc-HYNIC-TOC2to a clinical setting.

A model is proposed to explain transgranular-stress corrosion cracking (T-SCC) in face-centered cubic (fcc) materials. Crack propagation is shown to be anisotropic, in that growth near {110} < 001> is discontinuous due to crack arrest by dislocation blunting whereas growth away from this growth orientation is continuous. For the former case, renucleation of arrested cracks involves active dissolution of shear bands at the crack tip, which changes the stress state at Lomer-Cottrell locks, causing them to fail by cleavage. Once the crack is nucleated, its instantaneous macroscopic crack-growth velocity is considered to be comprised of multiple nucleation of microcracks with intervening arrests. This microcracking results from the interaction of the stress fields from neighboring cracks which are forming simultaneously, the crack-opening constraint due to ligaments which act as “bridges” behind the crack front, and the localized dissolution at the microcrack tip which affectsKICand leads to the “cobblestone” appearance. Experimental evidence and theoretical considerations are presented to support the model. The system studied was Cu-25 at. pct Au in 0.6 M NaCl solution at potentials between 300 and 400 mV (sce), which precludes hydrogen embrittlement.

Clear cell renal cell carcinoma (ccRCC) is a type of malignant tumor of the urinary system. The renal vein or vena cava thrombus can be found in a subset of ccRCC patients in whom it leads to worse prognosis. However, the protein expression profile and molecular features of ccRCC thrombus remain largely unclear. Here, a comparative proteomic analysis was performed using the 2D-LC-MS strategy for the thrombus-tumor-normal tissue triples of 15 ccRCC patients. Statistical analysis, GO enrichment analysis, protein-protein interaction network construction, and mRNA-based survival analysis were used to interpret the proteomic data. Three dysregulated proteins, GGT5 (gamma-glutamyl transferase 5), KRT7 (keratin 7) and CFHR1 (complement factor H related 1), were analyzed using western blot (WB) and immunohistochemistry (IHC) to validate the reliability of the proteomic analysis. The result of this analysis revealed 251 dysregulated proteins, which could be divided into 11 clusters depending on the changing trends, among the thrombus, tumor, and normal tissues. Several pathways and regulation networks were found to be associated with the thrombus, and some dysregulated proteins showed potential values for prognosis prediction. WB and IHC results were in accordance with the proteomic results, further validating the reliability of this study. In conclusion, our findings provide an overview of the thrombus at the molecular level as well as valuable information for further pathological studies or research on biomarkers and therapeutic targets.

The histopathological subtype of lung adenocarcinoma (LUAD) is closely associated with prognosis. Micropapillary or solid predominant LUAD tends to relapse after surgery at an early stage, whereas lepidic pattern shows a favorable outcome. However, the molecular mechanism underlying this phenomenon remains unknown. Here, we recruited 31 lepidic predominant LUADs (LR: low-risk subtype group) and 28 micropapillary or solid predominant LUADs (HR: high-risk subtype group). Tissues of these cases were obtained and label-free quantitative proteomic and bioinformatic analyses were performed. Additionally, prognostic impact of targeted proteins was validated using The Cancer Genome Atlas databases (n = 492) and tissue microarrays composed of early-stage LUADs (n = 228). A total of 192 differentially expressed proteins were identified between tumor tissues of LR and HR and three clusters were identified viahierarchical clustering excluding eight proteins. Cluster 1 (65 proteins) showed a sequential decrease in expression from normal tissues to tumor tissues of LR and then to HR and was predominantly enriched in pathways such as tyrosine metabolism and ECM-receptor interaction, and increased matched mRNA expression of 18 proteins from this cluster predicted favorable prognosis. Cluster 2 (70 proteins) demonstrated a sequential increase in expression from normal tissues to tumor tissues of LR and then to HR and was mainly enriched in pathways such as extracellular organization, DNA replication and cell cycle, and high matched mRNA expression of 25 proteins indicated poor prognosis. Cluster 3 (49 proteins) showed high expression only in LR, with high matched mRNA expression of 20 proteins in this cluster indicating favorable prognosis. Furthermore, high expression of ERO1A and FEN1 at protein level predicted poor prognosis in early-stage LUAD, supporting the mRNA results. In conclusion, we discovered key differentially expressed proteins and pathways between low-risk and high-risk subtypes of early-stage LUAD. Some of these proteins could serve as potential biomarkers in prognostic evaluation.