A sensitive method for the determination of mexiletine and lidocaine using surfactant-assisted dispersive liquid-liquid microextraction coupled with capillary electrophoresis was developed. Triton X-100 and dichloromethane were used as the dispersive agent and extraction solvent, respectively. After the extraction, mexiletine and lidocaine were analyzed using capillary electrophoresis with ultraviolet detection. The detection sensitivity was further enhanced through the use of field-amplified sample stacking. Under optimal extraction and stacking conditions, the calibration curves were linear over a concentration range of 0.05-1.00 μM for mexiletine and 0.03-1.00 μM for lidocaine. The limits of detection (signal-to-noise ratio of 3) were 0.01 and 0.01 μM for mexiletine and lidocaine, respectively. An approximately 1141- to 1250-fold improvement in sensitivity was observed for the two analytes compared with the injection of a standard solution without the surfactant-assisted dispersive liquid-liquid microextraction and field-amplified sample stacking procedures. This developed method was successfully applied to the determination of mexiletine and lidocaine in human urine and serum samples. Both precision and accuracy for urine and serum samples were less than 8.7 and 6.7%, respectively. The recoveries of the two analytes from urine and serum samples were 54.7-64.9% and 16.1-56.5%, respectively.

No abstract

In this work, Ag-doped HfO2-based resistive random access memory (RRAM) with high on-off ratio, low-power consumption and forming-free properties was investigated. We propose the fabrication flow of the RRAM with via-hole structure. After doping Ag into HfO2 as the switching layer, the devices could execute resistive switching without a high-voltage forming process. The conduction mechanism was subsequently validated by a current fitting analysis. Electric field simulation was also utilized to observe the electric field distribution and finally a physical model was proposed to provide an explanation for the formation and dissolution of the filament.

The diamond C͑001͒ surface has been observed by scanning tunneling microscopy ͑STM͒ to dimerize similar to Si͑001͒. STM also observed the formation of single-and double-layer steps in this surface. Using an ab initio multicenter molecular-dynamics method, our calculations for the relaxed geometries and the corresponding energetics of these steps show that the detailed structures of these steps are intrinsically different from that of type-A and type-B steps of Si͑001͒ obtained by Chadi. These differences can be attributed to the preference of carbon atoms near the edges of the lower terrace to form dimers with strong mixed single and double bonds rather than to rebond to the edge atoms. ͓S0163-1829͑98͒05127-3͔Studies of homoexpitaxial diamond films by chemical vapor deposition have recently been made toward the growth of high-quality single-crystalline diamond films and the understanding of growth mechanism. Morphological studies using scanning tunneling microscopy ͑STM͒ and low-energy electron diffraction have found (2ϫ1) reconstruction and singlelayer steps on the homoepitaxial diamond C͑001͒ films. 1-7 Double-layer steps with step edges parallel to the dimer rows have also been observed in coexistence with single-layer steps when the diamond ͑001͒ surface was annealed in hydrogen plasma. 6-10 Following Chadi, 11 we shall denote single-layer steps S A and S B and double-layer steps D A and D B as the steps with step edges that run parallel and perpendicular to the dimer rows on the upper terraces, respectively. STM measurements found that step edges of S A were almost straight, whereas those of S B were ragged with many kinks. These observations suggested that S A steps have a formation energy lower than that of S B similar to Si͑001͒ grown by molecular beam epitaxy. The occurrence of D A steps and the absence of D B steps in STM observations suggested that the formation energy of the D A step is lower than that of the D B step in contrast to Si͑001͒, which led Kuang et al. 10 to suggest that the saturation of dangling bonds by hydrogen atoms present in the experimental environment might reverse the preference of D B .To understand these STM observations, we have carried out this study using the ab initio multicenter molecular dynamics ͑MD͒ method of Sankey and Niklewski. 12 This method is based on the norm-conserving pseudopotential method 13,14 with s, p x , p y , and p z local-orbital basis set and has been shown to work well for semiconductor systems 15-18 and diamond. 19 The theoretical lattice constant of 3.7 Å ͑about 3.7% error from the experimental value 20 of 3.567 Å͒ determined by bulk diamond calculations using a cubic unit cell and repeated five-and six-layer slab ͑or supercell͒ models are used for S A and S B steps. For D A and D B steps, only repeated six-layer-slab models are used. The need for choosing both five-and six-layer supercell models for S A and S B steps will be described later. We have chosen a (11ϫ2) unit cell for all steps. For S A and S B steps, twelve surface atoms per unit cell f...

Colorectal cancer is the leading cause of cancer-associated morbidity and mortality worldwide. One of the causes of developing colorectal cancer is untreated colon adenomatous polyps. Clinically, polyps are detected in colonoscopy and the malignancies are determined according to the biopsy. To provide a quick and objective assessment to gastroenterologists, this study proposed a quantitative polyp classification via various image features in colonoscopy. The collected image database was composed of 1991 images including 1053 hyperplastic polyps and 938 adenomatous polyps and adenocarcinomas. From each image, textural features were extracted and combined in machine learning classifiers and machine-generated features were automatically selected in deep convolutional neural networks (DCNN). The DCNNs included AlexNet, Inception-V3, ResNet-101, and DenseNet-201. AlexNet trained from scratch achieved the best performance of 96.4% accuracy which is better than transfer learning and textural features. Using the prediction models, the malignancy level of polyps can be evaluated during a colonoscopy to provide a rapid treatment plan.