K. E. Akdogan

Analysis of direct motion measurement system for design of above knee prosthesis

In this study, motion measurement systems (MMS) are designed and analyzed in the laboratory for the necessary use in the development of any electronic above knee prosthesis (AKP). A low cost 2D image based MMS is one of the designs in order to analyze gait dynamics and provide reliable reference base for the comparison with direct MMSs which are also designed subsequently. Three direct MMSs using several combinations with accelerometers and/or gyroscopes are configured in the experimental set up and the records are compared to those obtained from image based MMS. The virtual sensor method showed the best accuracy and the approach using the combination of gyroscope and accelerometer had a lower accuracy while the calculations based on single accelerometers provide the worst performance in the relative knee angle estimations. It is recommended that MMSs employing virtual sensor and assembly of accelerometer and gyroscope are more reliable platforms for AKP designs.

Design of semi active knee joint with magnetorheological (MR) damper

In this study, electronic above knee prosthesis using magnetorheological (MR) cylinder is developed. The prosthesis structure having a proper knee mechanism is used for MR cylinder damping produced for experimental studies. Movement of knee joint originated from hip is damped with these cylinders and through a control algorithm running on microcontroller the system achieves the motion closer to the natural movement of leg. The performance of MR cylinder in the prosthesis are examined while a healthy subject wearing a custom designed socket mounted to prosthesis walks interchangeably at predetermined speed. According to the results obtained by the image based Motion Analysis System (MAS) the walk with the prosthesis using the MR damper is closer to the natural gait in terms of gait cycle duration and phases. On the other hand, in terms of the required maximum knee angle, the prosthesis with pneumatic damper performs better when the natural gait cycle is considered.

Simulations of knee angle control in dynamical Gait model for above knee prosthesis

Electronic above knee prosthesises need control methods and stratagies in order to conrol gait in an adaptive manner to user and enviroment. In this study, by using Proportional-Derivative (PD) and Computed Torque (CT) methods, knee control is analyzed. During gait, by modelling the movement of gait as two joint robot arm, dynamic equations of movement can be obtained. By applying artificial torques to dynamic equations, the results of simulation of gait without control is observed. Based on the problems observed, together with PD and CT methods, simulations are carried out for knee joint control. It is observed that the results of the simulations of controlled gait phases of gait dynamic model is close to the real data and follow the reference signal.

Investigation of optimum temporal update periods for regional TEC monitoring

Monitoring of the ionospheric variability is necessary for improving the performance of communication, navigation and positioning systems. Total Electron Content (TEC) is an important parameter since it can be used to analyze the variability of the ionosphere. Due to sparse distribution of samples, TEC can not be directly obtained everywhere on the ionosphere and accurate interpolation methods are needed for generating TEC maps. The optimum temporal update period (TUP) for regional Total Electron Content (TEC) monitoring has to be determined for capturing significant variations. In this study, the wide sense stationary (WSS) period is investigated whether it is possible to use the estimated WSS periods as the TUPs of the regional TEC maps. The sliding window statistical analysis method is used to estimate the WSS period. The WSS periods are compared with the difference between the maps obtained by Ordinary Kriging (OK) method. It is observed that the maximum WSS periods are following the length of time delays between the TEC maps at which the first significant difference occurs. The minimum WSS periods have to be used if the small variations are needed to be monitored within the day of interest. The WSS periods can be used as the TUPs for the regional TEC monitoring and can be adaptively adjusted.

Gait measurement using Radial Basis Function Networks and comparison of its performance with analytical methods

In this study, Radial Basis Function Networks (RBFN), a neural networks structure, is investigated to employ in sensor based motion measurement system of an electronic above knee prosthesis that ensures above knee amputees walk at varying speed. Having gait measurements of a custom designed image based measurement system as a reference, motion measurement performance of RBFN is compared with the performance of analytical methods on basis of knee angle measurements. Accuracy of knee angle estimations of RBFN is observed to be better than the accuracy of analytical methods using only accelerometers, other one using combination of accelerometer and gyroscope and another one called virtual sensor method. One advantage over analytical methods that RBFN offers is the possibility of decreasing number of sensors required to estimate knee angle. Even in case of single accelerometer attached on shank where lowest performance in knee angle estimation is observed, accuracy of RBFN is higher than the one analytical methods provide. From results of experiments, it is found that gyroscopes affects performance more than accelerometers do in knee angles estimation of RBFN.

Regional Ionosphere Mapping by using Neural Networks

Regional ionosphere mapping is necessary to understand the underlying characteristics of ionosphere. In this paper, multilayer perceptrons and radial basis networks, as chosen neural network methods, are used to obtain a map of ionosphere based on a number of given (Total Electron Content) TEC values. For varying number of given data and locations, their performances are analyzed and compared. Their performances are also tested for artificially formed data points and they are applied to real data from GPS stations and IRI.

Application of conformal transformation on the analysis of forward problems in electrical impedance tomography

Electrical impedance tomography (EIT) shows the conductivity distribution through the cross-section of a body part. It is reported that the image reconstruction is distorted considerably when the boundary shape is considered to be more elliptical than circular in a thorax model. The elliptic geometry model, which constructs more general structure, reduces the amount of problems and ensures more accurate results about perceiving inhomogeneities. In this study, the forward problem of EIT is analyzed under elliptical and circular frameworks by using an analytical and finite element method. The analytical solution defined for elliptic geometry is improved by applying the conformal transformation to the forward problem.

Analysis of forward problem for elliptical geometry in EIT by using analytical and finite element methods

Electrical impedance tomography (EIT) is a technique that computes the cross-sectional impedance distribution within the body. It is reported that the image reconstruction is distorted considerably when the boundary shape is considered to be more elliptical than circular in thorax model. In this paper, the forward problem of EIT is analyzed by analytical and verified by finite element model (FEM) for both circular and elliptic frames