Multilayer Diamond Coatings Applied to Micro-End-Milling of Cemented Carbide

Cobalt-cemented carbide micro-end mills were coated with diamond grown by chemical vapor deposition (CVD), with the purpose of micro-machining cemented carbides. The diamond coatings were designed with a multilayer architecture, alternating between sub-microcrystalline and nanocrystalline diamond layers. The structure of the coatings was studied by transmission electron microscopy. High adhesion to the chemically pre-treated WC-7Co tool substrates was observed by Rockwell C indentation, with the diamond coatings withstanding a critical load of 1250 N. The coated tools were tested for micro-end-milling of WC-15Co under air-cooling conditions, being able to cut more than 6500 m over a period of 120 min, after which a flank wear of 47.8 μm was attained. The machining performance and wear behavior of the micro-cutters was studied by scanning electron microscopy and energy-dispersive X-ray spectroscopy. Crystallographic analysis through cross-sectional selected area electron diffraction patterns, along with characterization in dark-field and HRTEM modes, provided a possible correlation between interfacial stress relaxation and wear properties of the coatings. Overall, this work demonstrates that high adhesion of diamond coatings can be achieved by proper combination of chemical attack and coating architecture. By preventing catastrophic delamination, multilayer CVD diamond coatings are central towards the enhancement of the wear properties and mechanical robustness of carbide tools used for micro-machining of ultra-hard materials.     

The effect of delivering the chemokine SDF-1α in a matrix-bound manner on myogenesis

Several chemokines are important in muscle myogenesis and in the recruitment of muscle precursors during muscle regeneration. Among these, the SDF-1α chemokine (CXCL12) is a potent chemoattractant known to be involved in muscle repair. SDF-1α was loaded in polyelectrolyte multilayer films made of poly(l-lysine) and hyaluronan to be delivered locally to myoblast cells in a matrix-bound manner. The adsorbed amounts of SDF-1α were tuned over a large range from 100 ng/cm² to 5 μg/cm², depending on the initial concentration of SDF-1α in solution, its pH, and on the film crosslinking extent. Matrix-bound SDF-1α induced a striking increase in myoblast spreading, which was revealed when it was delivered from weakly crosslinked films. It also significantly enhanced cell migration in a dose-dependent manner, which again depended on its presentation by the biopolymeric film. The low-crosslinked film was the most efficient in boosting cell migration. Furthermore, matrix-bound SDF-1α also increased the expression of myogenic markers but the fusion index decreased in a dose-dependent manner with the adsorbed amount of SDF-1α. At high adsorbed amounts of SDF-1α, a large number of Troponin T-positive cells had only one nucleus. Overall, this work reveals the importance of the presentation mode of SDF-1α to emphasize its effect on myogenic processes. These films may be further used to provide insight into the role of SDF-1α presented by a biomaterial in physiological or pathological processes.     

Neural architecture design based on extreme learning machine

Selection of the optimal neural architecture to solve a pattern classification problem entails to choose the relevant input units, the number of hidden neurons and its corresponding interconnection weights. This problem has been widely studied in many research works but their solutions usually involve excessive computational cost in most of the problems and they do not provide a unique solution. This paper proposes a new technique to efficiently design the MultiLayer Perceptron (MLP) architecture for classification using the Extreme Learning Machine (ELM) algorithm. The proposed method provides a high generalization capability and a unique solution for the architecture design. Moreover, the selected final network only retains those input connections that are relevant for the classification task. Experimental results show these advantages.     

Modulating the release kinetics through the control of the permeability of the layer-by-layer assembly: a review

The layer-by-layer (LbL) self-assembly technique has emerged as a simple and versatile method for coating biological and non-biological templates for various biomedical applications. A promising avenue of this technique lies in the encapsulation of drugs and other biological substances for controlled release. Fundamental studies of LbL assembly on flat surfaces have provided a sound understanding of film deposition theory and its pertinence to ionic and molecular transport and diffusion through polyelectrolyte multilayer (PEM) films. However, there is a lack of information on the permeability of three-dimensional PEM shell systems. In either PEM films or shells, it has been shown that drug release is a function of the ionic strength, pH and/or multilayer thickness. This report aims to provide an overview of the physicochemical parameters affecting the permeability of two- and three-dimensional multilayer shells, including ionic strength, layer number and pH. Furthermore, their synergic effect on loading and release of biologically active molecules from LbL multilayers are discussed.     

Development and validation of artificial intelligence models for preoperative prediction of inferior mesenteric artery lymph nodes metastasis in left colon and rectal cancer

BackgroundDissection of lymph nodes at the roots of the inferior mesenteric artery (IMAN) should be offered only to selected patients at a major risk of developing IMAN involvement. The aim of this study is to present the first artificial intelligence (AI) models to predict IMAN metastasis risk in the left colon and rectal cancer patients.MethodsA total of 2891 patients with descending colon including splenic flexure, sigmoid colon and rectal cancer undergoing major primary surgery and IMAN dissection were included as a study cohort, which was then split into a training set (67%) and a testing set (33%). Feature selection was conducted using the least absolute shrinkage and selection operator (LASSO) regression model. Seven AI algorithms, namely Support Vector Machine (SVM), Logistic Regression (LR), Extreme Gradient Boosting (XGB), Light Gradient Boosting (LGB), Decision Tree Classifier (DTC), Random Forest (RF) classifier, and Multilayer Perceptron (MLP), as well as traditional multivariate LR model were employed to construct predictive models. The optimal hyperparameters were determined with 5 fold cross-validation. The predictive performance of models and the expert surgeon was assessed and compared in the testing set independently.ResultsThe IMAN involvement incidence was 4.6%. The optimal set of features selected by LASSO included 10 characteristics: neoadjuvant treatment, age, synchronous liver metastasis, synchronous lung metastasis, signet ring adenocarcinoma, neural invasion, lymphovascular invasion, CA199, endoscopic obstruction, T stage evaluated by MRI. The most accurate model derived from MLP showed excellent prediction power with area under the receiver operating characteristic curve (AUROC) of 0.873 and produced 81.0% recognition sensitivity and 82.5% specificity in the testing set independently. In contrast, the judgment of IMAN metastasis by expert surgeon yield rather imprecise and unreliable results with a significantly lower AUROC of 0.509. Additionally, the proposed MLP had the highest net benefits and the largest reduction of unnecessary IMAN dissection without the cost of additional involved IMAN missed.ConclusionMLP model was able to maintain its prediction accuracy in the testing set better than other models and expert surgeons. Our MLP model could be used to help identify IMA nodal metastasis and to select candidates for individual IMAN dissection.     

Speed limits in the cerebellum: constraints from myelinated and unmyelinated parallel fibers

Cerebellar parallel fibers are among the thinnest known vertebrate axons and represent an extreme anatomical adaptation. Until now a systematic examination of their properties across species has not been carried out. We used transmission electron microscopy and light microscopy to compare parallel fibers in mammals of different brain sizes. From mouse to macaque, the average unmyelinated parallel fiber diameter was 0.2-0.3 microm, consistent with the idea that they are evolutionarily selected for compactness. Average unmyelinated parallel fiber diameter scaled up slightly with brain size, and across species the estimated total conduction time is 5-10 ms. However, these conduction times can vary by milliseconds, and unmyelinated PFs consume large amounts of energy per action potential. These functional disadvantages are overcome in myelinated parallel fibers, which we found in the deep regions nearest the Purkinje cell layer in marmoset, cat and macaque. These axons were 0.4-1.1 microm wide, have expected conduction times of 0.5-1.0 ms, and may convey fast feedforward inhibition via basket cells to Purkinje cells.     

Dry-phase technology for immunoassays

Dry-phase immunoassay technologies are reviewed. The systems described include the incorporation of homogeneous immunoassay technologies, such as SLFIA, ARIS, and enzyme-channeling methods, into paper strips. Also described are the different approaches to integrating ELISA methods into multizoned reagent devices that allow convenient automated or manual assays.     

Finite element modelling of forearm skin wrinkling

Background/purpose: Human skin is a complex multilayered material. Although there are many numerical models of skin in existence, which accurately simulate several of its complex mechanical characteristics, there are very few models that simulate wrinkling – a phenomenon common to all human skin. The purpose of this study was to develop a multilayer model of skin, which could simulate wrinkling more realistically than the existing models in the literature. Method: The skin on the volar forearm was wrinkled by displacing two tabs, which were attached to the skin surface towards each other. Eight volunteers were used in the experiment. The profiles of the wrinkles were measured using a customised laser profilometer system. The maximum range and average roughness of the wrinkle profiles were calculated. The multilayer model of skin consisted of the stratum corneum, dermis and underlying hypodermis. The constitutive equations for each layer were implemented into a finite element model and the wrinkling experiments were simulated. The results of the model were compared with the results of the experiments. Results: The size of the wrinkles predicted by the model fell within the range of the wrinkle sizes measured in the experiments. The maximum range and average roughness differed by 34% and 43% from the corresponding mean experimental results, respectively. The results show that the proposed three‐layer skin model simulates wrinkling more realistically than either a single or a two‐layer model. Conclusion: A three‐layer model of skin has been developed and validated under the harsh conditions of wrinkling. It has been shown that a three‐layer model better simulates the wrinkling of skin as compared with models of fewer layers. The model has several applications, including simulating skin aging and the design of more realistic artificial skin. The model can also be used to study the interaction of surgical scalpels or razors with human skin with a view to improving their design.