ExoMechHand prototype development and testing with EMG signals for hand rehabilitation.

Rehabilitation is a major requirement to improve the quality of life and mobility of patients with disabilities. The use of rehabilitative devices without continuous supervision of medical experts is increasing manifold, mainly due to prolonged therapy costs and advancements in robotics. Due to ExoMechHand's inexpensive cost, high robustness, and efficacy for participants with median and ulnar neuropathies, we have recommended it as a rehabilitation tool in this study. ExoMechHand is coupled with three different resistive plates for hand impairment. For efficacy, ten unhealthy subjects with median or ulnar nerve neuropathies are considered. After twenty days of continuous exercise, three subjects showed improvement in their hand grip, range of motion of the wrist, or range of motion of metacarpophalangeal joints. The condition of the hand is assessed by features of surface-electromyography signals. A Machine-learning model based on these features of fifteen subjects is used for staging the condition of the hand. Machine-learning algorithms are trained to indicate the type of resistive plate to be used by the subject without the need for examination by the therapist. The extra-trees classifier came out to be the most effective algorithm with 98% accuracy on test data for indicating the type of resistive plate, followed by random-forest and gradient-boosting with accuracies of 95% and 93%, respectively. Results showed that the staging of hand condition could be analyzed by sEMG signal obtained from the flexor-carpi-ulnaris and flexor-carpi-radialis muscles in subjects with median and ulnar neuropathies.

Deploying efficient net batch normalizations (BNs) for grading diabetic retinopathy severity levels from fundus images.

Diabetic retinopathy (DR) is one of the main causes of blindness in people around the world. Early diagnosis and treatment of DR can be accomplished by organizing large regular screening programs. Still, it is difficult to spot diabetic retinopathy timely because the situation might not indicate signs in the primary stages of the disease. Due to a drastic increase in diabetic patients, there is an urgent need for efficient diabetic retinopathy detecting systems. Auto-encoders, sparse coding, and limited Boltzmann machines were used as a few past deep learning (DL) techniques and features for the classification of DR. Convolutional Neural Networks (CNN) have been identified as a promising solution for detecting and classifying DR. We employ the deep learning capabilities of efficient net batch normalization (BNs) pre-trained models to automatically acquire discriminative features from fundus images. However, we successfully achieved F1 scores above 80% on all efficient net BNs in the EYE-PACS dataset (calculated F1 score for DeepDRiD another dataset) and the results are better than previous studies. In this paper, we improved the accuracy and F1 score of the efficient net BNs pre-trained models on the EYE-PACS dataset by applying a Gaussian Smooth filter and data augmentation transforms. Using our proposed technique, we have achieved F1 scores of 84% and 87% for EYE-PACS and DeepDRiD.

Cattaneo-Christov Double Diffusion (CCDD) on Sutterby Nanofluid with Irreversibility Analysis and Motile Microbes Due to a RIGA Plate.

In this article, a Riga plate is exhibited with an electric magnetization actuator consisting of permanent magnets and electrodes assembled alternatively. This Riga plate creates an electric and magnetic field, where a transverse Lorentz force is generated that contributes to the flow along the plate. A new study field has been created by Sutterby nanofluid flows down the Riga plate, which is crucial to the creation of several industrial advancements, including thermal nuclear reactors, flow metres, and nuclear reactor design. This article addresses the second law analysis of MHD Sutter by nanofluid over a stretching sheet with the Riga plate. The Cattaneo-Christov Double Diffusion heat and mass flux have been created to examine the behaviour of relaxation time. The bioconvection of motile microorganisms and chemical reactions are taken into consideration. Similarity transformations are used to make the governing equations non-dimensional ordinary differential equations (ODE's) that are subsequently solved through an efficient and powerful analytic technique, the homotopy analysis method (HAM). The effect of pertained variables on velocity, temperature, concentration, and motile microorganism distributions are elaborated through the plot in detail. Further, the velocity distribution enhances and reduces for greater value Deborah number and Reynold number for the two cases of pseudoplastic and dilatant flow. Microorganism distribution decreases with the augmented magnitude of Peclet number (Pe), Bioconvection Lewis number (Lb), and microorganism concentration difference number (ϖ). The entropy production distribution is increased for the greater estimations of the Reynolds number (ReL) and Brinkman parameter (Br). Two sets of graphical outputs are presented for the Sutterby fluid parameter. Finally, for the justification of these outcomes, tables of comparison are made with various variables.

Flexure and shear response of an impulsively loaded rigid-plastic beam by enhanced linear complementarity approach.

The linear complementarity approach has been utilized as a systematic and unified numerical process for determining the response of a rigid-plastic structure subjected to impulsive loading. However, the popular Lemke Algorithm for solving linear complementarity problems (LCP) encounters numerical instability issues whilst tracing the response of structures under extreme dynamic loading. This paper presents an efficient LCP approach with an enhanced initiation subroutine for resolving the numerical difficulties of the solver. The numerical response of the impulsively loaded structures is affected by the initial velocity profile, which if not found correctly can undermine the overall response. In the current study, the initial velocity profile is determined by a Linear Programming (LP) subroutine minimizing the energy function. An example of a uniform impulsively loaded simply supported beam is adduced to show the validity and accuracy of the proposed approach. The beam is approximated with bending hinges having infinite resistance to shear. Comparison of the numerical results to the available closed-form solution confirms the excellent performance of the approach. However, a subsequent investigation into a beam having the same support conditions and the applied loading, but with bending and shear deformation, results in numerical instability despite optimizing the initial velocity profile. Thus a more generic description of kinetics and kinematics is proposed that can further enhance the numerical efficiency of the LCP formulation. The ensuing numerical results are compared with the available close form solution to assess the accuracy and efficiency of the developed approach.

Fully developed Darcy-Forchheimer mixed convective flow over a curved surface with activation energy and entropy generation.

BACKGROUND: Mixed convection (forced+natural convection) is frequently observed in exceptionally high output devices where the forced convection isn't sufficient to dissipate all of the heat essential. At this point, consolidating natural convection with forced convection will frequently convey the ideal outcomes. Nuclear reactor technology and a few features of electronic cooling are the examples of these processes. Mixed convection problems are categorized by Richardson number (Ri), which is the ratio of Grashof number (for natural convection) and Reynolds number (for forced convection). For buoyancy or mixed convection the relative effect can be addressed by Richardson number. Typically, the natural convection is negligible when Richardson number is less than 0.1 (Ri < 0.1), forced convection is negligible when Richardson number is greater than 10 (Ri > 10) and neither is negligible when (0.1 < Ri < 10). It might be noticed that generally the forced convection is large comparative with natural convection except in case of remarkably low forced flow velocities. The current work gives significant insights regarding dissipative mixed convective Darcy-Forchheimer flow with entropy generation over a stretched curved surface. The energy equation is developed with respect to nonlinear radiation, dissipation and Ohmic heating (Joule heating). Binary reaction via activation energy is accounted. METHOD: Curvilinear transformations are utilized to change the nonlinear PDE's into ordinary ones. Computational outcomes are obtained via NDSolve MATHEMATICA. The results are computed and discussed graphically. RESULTS: Velocity decays for Forchheimer number. Entropy generation enhances for diffusion parameter and chemical reaction parameter. Concentration profile reduces chemical reaction parameter and enhances for activation parameter.

Magnetohydrodynamics (MHD) radiated nanomaterial viscous material flow by a curved surface with second order slip and entropy generation.

BACKGROUND: Magnetohydrodynamics or hydro-magnetics (MHD) is the study of dynamics in the presence of magnetic characteristics and impact of electrically conducting liquids which has a significant applications in engineering and biomedical sciences. Liquid metals, plasma, electrolytes and salt water are the examples of such magneto-fluids. MHD liquid flow in various geometries significant to engineering sciences is an interesting and noteworthy scientific area because of applications. The above applications of magnetohydrodynamics insist the engineers and analyst to develop new mathematical modeling in the field of fluid mechanics. Therefore, we considered electrical conducting viscous fluid flow over a curved surface with second order slip. The Buongiorno model is utilized in the modeling of flow problem with thermophoretic and Brownian diffusions. The effects of viscous dissipation, thermal radiation and Joule heating (Ohmic heating) is used in the modeling of energy equation. Homogeneous-heterogeneous reactions are further considered. The energy equation is modeled. METHOD: The nonlinear ODE's are obtained through utilization of appropriate transformations and numerical results are computed via NDSolve MATHEMATICA. RESULTS: Velocity field is decreasing function of first order slip parameter. Both Bejan number and entropy generation is upsurged versus heterogeneous reaction parameter.

Modeling and computational analysis of hybrid class nanomaterials subject to entropy generation.

BACKGROUND AND OBJECTIVE: Nanoliquids are dilute suspensions of nanoparticles with at least one of their principal dimensions smaller than 100 nm. Form literature, nanoliquids have been found to possess increased thermos-physical characteristics like thermal diffusivity, thermal conductivity, convective heat transport coefficients and viscosity associated to those of continuous phase liquids foe example oil, ethylene glycol and water. Nanoliquids have novel characteristics that make them possibly beneficial in numerous applications in heat transport like fuel cells, microelectronics, hybrid-powered engines, pharmaceutical processes, domestic refrigerator, engine cooling thermal management, chiller and heat exchanger. The above applications of nanofluids/hybrid nanofluids insist the researchers and engineers to develop new methodologies and technique in the field of heat transport. Therefore, we have considered mixed convective flow hybrid nanomaterial over a convectively heated surface of disk. Flow nature is discussed due to stretchable rotating surface of disk. Applied magnetic field is accounted. Ohmic heating and dissipation effects are utilized in the modeling of energy expression. Total entropy rate is calculated. METHODS: Suitable transformation leads to ordinary differential equations. Shooting method is implemented for numerical outcomes. Comparative analysis is made for the present result with published ones. RESULTS: The effects of key parameters like magnetic parameter, mixed convection variable and Eckert and Biot numbers on the dimensionless velocity, surface drag force, temperature, (heat transfer rate) Nusselt number and entropy rate are discussed in detail and presented graphically. Furthermore, the outcomes demonstrate that velocity of liquid particles decline against magnetic parameter. Temperature and associated layer upsurge versus magnetic parameter and Eckert number. Skin friction coefficient (drag force) improves through higher values of stretching and magnetic variables. Heat transfer rate is more for higher Eckert number and magnetic parameter. Entropy rate is also enhances against Eckert number and Brickman number. CONCLUSIONS: Magnitude of surface drag force increases for higher values of stretching and magnetic variables. Magnitude of heat transfer rate is more when magnetic variable and Eckert number attain the maximum values. Brinkman number is used to decrease the entropy rate. Furthermore, velocity and temperature show contrast behavior versus magnetic parameter i.e., velocity of fluid particles decreases.