Objective To present new classification methods of knee osteoarthritis (KOA) using machine learning and compare its performance with conventional statistical methods as classification techniques using machine learning have recently been developed.Methods A total of 84 KOA patients and 97 normal participants were recruited. KOA patients were clustered into three groups according to the Kellgren-Lawrence (K-L) grading system. All subjects completed gait trials under the same experimental conditions. Machine learning-based classification using the support vector machine (SVM) classifier was performed to classify KOA patients and the severity of KOA. Logistic regression analysis was also performed to compare the results in classifying KOA patients with machine learning method.Results In the classification between KOA patients and normal subjects, the accuracy of classification was higher in machine learning method than in logistic regression analysis. In the classification of KOA severity, accuracy was enhanced through the feature selection process in the machine learning method. The most significant gait feature for classification was flexion and extension of the knee in the swing phase in the machine learning method.Conclusion The machine learning method is thought to be a new approach to complement conventional logistic regression analysis in the classification of KOA patients. It can be clinically used for diagnosis and gait correction of KOA patients.
We developed a cell voltage model as a function of the current density response of the direct methanol fuel cell ͑DMFC͒. The impact on cell performance by methanol crossover, which lowers the cell performance, is discussed. We employ a srmethanol activity term to take into account the methanol crossover effect. A methanol activity equation is derived from the Helmholtz energy of mixing. In this study, we use a simplified expression of the Helmholtz energy of mixing and correct the mathematical approximation defect using the fractional form. To take the specific interaction into account, a secondary lattice concept, perturbation to a fixed reference system, is introduced. A cell voltage equation with reasonable boundary conditions, including a methanol crossover effect that plays a major role in determining the cell voltage of DMFC, is developed. It can also be found that the methanol crossover increases with an increase in membrane chain length.The increasing use of electricity and concern about the environmental consequences of fossil fuel use has brought a great amount of attention on fuel cells as a potential technology to replace the internal combustion engine. The proton exchange membrane fuel cell ͑PEMFC͒ and direct methanol fuel cell ͑DMFC͒ are presently considered as two promising types of fuel cells for day-to-day applications. The DMFC is an attractive power source due to its low operating temperature and high theoretical energy density. In the DMFC, methanol is directly fed into the fuel cell. Compared to PEMFC, DMFC has further advantages of transportation, storage, and compact cell design. However, at present, there are several challenging problems to the wide application of DMFC. These consist mainly of the sluggish reaction of methanol oxidation at the anode electrode, methanol crossover through the polymer membrane, and severe cathode flooding. The cell performance is limited by anode kinetics due to its low exchange current density and high Tafel slope. 1 Furthermore, methanol crossover causes lower open-circuit voltage and wastes fuel. Water management greatly influences the cathode performance. 2 Methanol crossover is due to permeability through the membrane, caused by the diffusion and the electro-osmotic drag of methanol from anode to cathode. The methanol movement across the membrane causes a mixed potential due to its oxidation and, consequently, a decrease in cell performance. Methanol crossover in DMFC has been extensively studied and shown to depend on a number of factors. 3 The most important ones are the membrane permeability/thickness, the concentration of methanol in the fuel feed, the operating temperature, and the performance of the anode itself. The membrane is a very important factor regarding the methanol crossover problem. Thinner membranes give lower resistances in the cell but tend to have a higher permeability for liquid methanol. 4 The crossover effect is dependent on the methanol concentration in the feed. An optimum concentration of methanol was considered to be ϳ1-2 M me...
Molecular dynamics simulations are used to study highly cross‐linked epoxy networks comprised of furanyl epoxy monomer, 2,5‐bis[(2‐oxiranylmethoxy)methyl]‐furan (BOF), that is cross‐linked by two furanyl amine hardeners, 5,5'‐methylenedifurfurylamine (DFDA) and 5,5'‐ethylidenedifurfirylamine (CH3‐DFDA). Important properties of these fully furan‐based systems, including room temperature density, glass transition temperature, and Young's modulus are found to agree with previous experimental results. We also compare the simulated and experimental values of four fully furan‐based thermosetting materials to those using the conventional resin diglycidyl ether of bisphenol A (DGEBA) cured with the two furanyl hardeners. Our simulation results predict a slight decrease in density and Young's modulus, but no impact on the glass transition temperature, upon adding the methyl group in DFDA. Detailed analyses of the MD trajectories reveal the underlying mechanisms responsible for the observed structure/property relations, which center on the lack of collinear covalent bonds in the BOF molecular structure. © 2016 Wiley Periodicals, Inc. J. Polym. Sci., Part B: Polym. Phys. 2017, 55, 285–292
Phase behaviors of polymer solutions are estimated using a combination of thermodynamic models and molecular simulation technique. In general, many parameters of binary systems are determined by fitting a thermodynamic model with experimental data. In this study, we obtained all parameters using molecular simulation. To take the specific interaction into account, we assume that it only occurs between a solvent molecule and a specific group. Our results show that the theoretical treatment accounting for the specific interaction gives more accurate predictions than those without consideration of specific interaction. Also, our approach describes the phase equilibria of various polymer solutions over the entire concentration remarkably well.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.