DTV-CNN: Neural network based on depth and thickness views for efficient 3D shape classification.

Fast and effective algorithms for deep learning on 3D shapes are keys to innovate mechanical and electronic engineering design workflow. In this paper, an efficient 3D shape to 2D images projection algorithm and a shallow 2.5D convolutional neural network architecture is proposed. A smaller convolutional neural network (CNN) model is achieved by information enrichment at the preprocessing stage, i.e. 3D geometry is compressed into 2D "thickness view" and "depth view". Fusing the depth view and thickness view (DTV) from the same projection view into a dual-channel grayscale image, can improve information locality for geometry and topology feature extraction. This approach bridges the gap between mature image deep learning technologies to the applications of 3D shape. Enhanced by several essential scalar geometry properties and only 3 projection views, a mixed CNN and multiple linear parameter (MLP) neural network model achives a validation accuracy of 92 % for ModelNet10 mesh-based dataset, while the training time is one order of magnitude less than the original multi-view CNN approach. This study also creates new 3D shape datasets from 2 open source CAD projects. Higher validation accuracy is obtained for realistic CAD datasets, i.e. 97 % for FreeCAD's mechanical part library and 95 % for KiCAD electronic part library. The training cost reduces to tens of minutes on a laptop CPU, given the smaller input data size and shallow neural network design. It is expected that this approach can be adapted for other machine learning scenarios involved in CAD geometry.

A Multi-Objectives Genetic Algorithm Based Predictive Model and Strategy Optimization during SLM Process.

Selective laser melting (SLM) process was optimized in this work using multi-objectives genetic algorithm. Process parameters involved in the printing process have an obvious impact on the quality of the printed parts. As the relationship between process parameters and the quality of different parts are complex, it is quite essential to study the effect of process parameter combination. In this work, the impact of four main process parameters, including defocusing amount, laser power, scan speed and layer thickness, were studied on overhanging surface quality of the parts with different inner structures. A multiple-factor and multiple-level experiment was conducted to establish a prediction model using regression analysis while multi-objective genetic algorithm was also employed here to improve the overhanging surface quality of parts with different inner shapes accordingly. The optimized process parameter combination was also used to print inner structure parts and compared with the prediction results to verify the model we have obtained before. The prediction results revealed that sinking distance and roughness value of the overhanging surface on a square-shape inner structure can reduce to 0.017 mm and 9.0 µm under the optimal process parameters combination, while the sinking distance and roughness value of the overhanging surface on a circle-shape inner structure can decrease to 0.014 mm and 10.7 µm under the optimal process parameters combination respectively. The testing results showed that the error rates of the prediction results were all within 10% in spite of random powder bonding in the printing process, which further proved the reliability of the previous results.

Real-Time and High-Fidelity Tracking of Lysosomal Dynamics with a Dicyanoisophorone-Based Fluorescent Probe.

The development of high-performance probes that can visualize and track the dynamic changes of lysosomes is very important for the in-depth study of lysosomes. Herein, we report that a dicyanoisophorone-based probe (named DCIP) can be used for high-fidelity imaging of lysosomes and lysosomal dynamics. DCIP can be easily prepared and shows strong far-red to near-infrared emissions centered at 653 nm in water with a huge Stokes shift (224 nm), high quantum yield (Φ = 0.15), high pKa value (∼8.79), and good biocompatibility. DCIP also shows good cell permeability and can label lysosomes rapidly with bright fluorescence without a time-consuming washing process before imaging. DCIP also possesses good photostability and negligible background, making it effective for long-term and high spatiotemporal resolution (0.44 s of exposure) imaging of lysosomes. Moreover, DCIP achieved high-fidelity tracking of lysosomal dynamics at an extremely low concentration (1 nM). Finally, we also demonstrated that DCIP could real-time track the interactions of lysosomes with other organelles (damaged mitochondria as a model) and image the drug-escape processes from lysosomes. All of the results show that DCIP holds broad prospects in lysosome-related research.

Real-time tracking lysosomal pH changes under heatstroke and redox stress with a novel near-infrared emissive probe.

Lysosomes are important subcellular organelles with acidic pH. The change of lysosomal pH can affect the normal function and activity of cells. To conveniently detect and visualize lysosomal pH changes, we designed herein a novel fluorescent probe NIR-Rh-LysopH. The probe is based on a Rhodamine 101 derivative, which was modified to include a fused tetrahydroquinoxaline ring to obtain near-infrared fluorescence and a methylcarbitol moiety to locate the lysosome. Based on the proton-induced spirolactam ring-opening mechanism, NIR-Rh-LysopH showed rapid, selective, sensitive, and reversible near-infrared fluorescence responses around 686 nm (Stokes shift 88 nm) with a pKa value of 5.70. From pH 7.4 to 4.0, about 285 folds of fluorescence enhancement was observed. Cell experiments showed that NIR-Rh-LysopH has low cytotoxicity and excellent lysosome-targeting ability. Moreover, NIR-Rh-LysopH was applied successfully to track lysosomal pH changes induced by drugs (such as chloroquine and dexamethasone), heatstroke, and redox stress. Thus, NIR-Rh-LysopH is very promising for conveniently tracking lysosomal pH changes and studying the related life processes.

In Vivo Imaging and Tracking Carbon Monoxide-Releasing Molecule-3 with an NIR Fluorescent Probe.

As a water-soluble carbon monoxide-releasing molecule, CORM-3 is widely used as a CO donor to study CO in the life system. CORM-3 can also replace gaseous CO as a therapeutic drug molecule to reveal the physiological and pathological effects of CO in life. Therefore, it is of great importance to visualize and track CORM-3 in the life system. We develop herein a near-infrared (NIR) fluorescent probe CORM3-NIR that can detect CORM-3 both in living cells and in vivo effectively. The probe is based on the unique fluorescent QCy7 and uses a 4-nitrobenzyl group to trap CORM-3, and importantly, it shows good water solubility and responds rapidly, selectively, and sensitively to CORM-3, releasing QCy-7 and producing distinct colorimetric and significant NIR fluorescence change signals at 743 nm. The Stokes shift is up to 81 nm. The probe is also able to detect CORM-3 ratiometrically with fluorescence at 743 and 600 nm. Besides, with low cytotoxicity, the probe also shows good NIR fluorescence bioimaging ability for CORM-3 in live cells and mice, which indicates that CORM3-NIR is an effective probe for tracking and studying CORM-3 in the life system.

A novel reaction-based fluorescence probe for rapid imaging of HClO in live cells, animals, and injured liver tissues.

Hypochlorous acid (HClO) is an important bioactive molecule, playing vital roles in a large number of physiological and pathological processes. Abnormal concentration of HClO in vivo has close contact with many diseases including inflammatory diseases and cancer. For bioimaging HClO, a new colorimetric and turn-on near-infrared (NIR) fluorescence probe (DDAO-ClO) was designed and synthesized in this work through a specific reaction of HClO with dimethylthiocarbamate. DDAO-ClO proved to show distinct and highly selective colorimetric and NIR fluorescence responses for HClO with fast response time (<3 s) and high sensitivity (LOD = 7.3 nM) in vitro. After confirming the excellent in vitro sensing ability, imaging HClO with DDAO-ClO in living HeLa cells, MCF-7 cells, zebrafish, and mice was all successfully demonstrated. And with this probe, it was further discovered that more endogenous HClO was produced in injured mice liver tissues, which demonstrates that DDAO-ClO not only is effective for in vivo detection of HClO but also has a broad application prospect.

NIR fluorescent probe based on a modified rhodol-dye with good water solubility and large Stokes shift for monitoring CO in living systems.

Because of the biological importance of CO, the development of effective probes for the detection of CO in living systems is of great significance. In particular, near-infrared (NIR) fluorescence probes with good water solubility and large Stokes shift are indispensable tools for CO detection in vivo. However, such tools are extremely scarce. Herein, a modified rhodol dye was used to develop a novel NIR fluorescent probe (Rh-NIR-CO) for selective and sensitive detection of CO using the Pd0-mediated Tsuji-Trost reaction. This probe shows good water solubility and rapid CO detecting ability in aqueous buffer at pH 7.4, accompanied by distinct colorimetric and turn-on NIR emission changes at 676 nm with a large Stokes shift (135 nm) and low detection limit (37 nM). Moreover, NIR fluorescence imaging of CO in living cells, zebrafish, and living mice was successfully applied with this probe. These excellent results highlighted Rh-NIR-CO as a promising new tool for in vitro and in vivo detection of CO.

DNA cleavage mediated by copper superoxide dismutase via two pathways.

The known action of Cu, Zn superoxide dismutase (Cu(2)Zn(2)SOD) that converts O(2)(-) to O(2) and H(2)O(2) plays a crucial role in protecting cells from toxicity of oxidative stress. However, the overproduction of Cu(2)Zn(2)SOD does not result in increased protection but rather creates a variety of unfavorable effects, suggesting that too much Cu(2)Zn(2)SOD may be injurious to the cells. The present study examined the DNA cleavage activity mediated by a Cu(n)SOD that contains 1-4 copper ions, in order to obtain an insight into the aberrant copper-mediated oxidative chemistry in the enzyme. A high SOD activity was observed upon metallation of the apo-form of Cu(2)Zn(2)SOD with Cu(II), indicating that nearly all of the Cu(II) in the Cu(n)SOD is as active as the Cu(II) in the copper site of fully active Cu(2)Zn(2)SOD. Using a supercoiled DNA as substrate, significant DNA cleavage was observed with the Cu(n)SOD in the presence of hydrogen peroxide or mercaptoethanol, whereas DNA cleavage with free Cu(II) ions can occur only <5% under the same conditions. Comparison with other proteins shows that the DNA cleavage activity is specific to some proteins including the Cu(n)SOD. The steady state study suggests that a cooperative action between the SOD protein and the Cu(II)may appear in the DNA cleavage activity, which is independent of the number of Cu(II) in the Cu(n)SOD. The kinetic study shows that a two-stage reaction was involved in DNA cleavage. The effects of various factors including EDTA, radical scavengers, bicarbonate anion, and carbon dioxide gas molecules on the Cu(n)SOD-mediated DNA cleavage activity were also investigated. It is proposed that DNA cleavage occurs via both hydroxyl radical oxidation and hydroxide ion hydrolysis pathways. This work implies that any form of the copper-containing SOD enzymes (including Cu(2)Zn(2)SOD and its mutants) might have the DNA cleavage activity.