Fusion of Variables and Processes. Optimization in Prosthetics using Regression Models
- Author Ilea Mihai
- Co-Author Turnea Marius, Rotariu Mariana, Arotaritei Dragos, Filep Robert
- DOI
- Country : Romania
- Subject : Medical Biosciences
In this paper, we propose a new objective method for estimating the comfort in prosthetics involving a stump-cuff-cup optimisation, a linear correlation and fusion of measurement variables. The basic idea is that the person with prosthesis feels better (has a greater comfort and satisfaction) if a combination of personal sensory perceptions relative to comfort / discomfort are perceived with varying degrees of acceptability.The aim of this paper is to find an effective method to select an optimal combination for socket-cup material (the same type of prosthesis) of a proper choose for one type of prosthesis using customized on a specific patient. For modelling and simulation, the stump-prosthesis-cuff assembly is scanned and it is determined the set of points for each element. The elements are re-assembled into a 3D spatial reconstruction. The patient’s perception on the handicap of prosthetics functionally and sensorially is determined by the method of the questionnaire. We propose a linear regression model in the first phase, which takes into account all the variables that influence the patient’s comfort with transtibial prosthesis. It is considered that different combinations of cuff-cup materials can produce various levels of satisfaction. This assertion is validated experimentally and specified in this article when different patients, depending on the person, feel comfortable at the maximum level with different types of cuffs and different materials. It is noticeable that in the fusion of non-optimised variables, the coincidence values are at least as good as the best ones for the individual variable. We proposed a method and application for objective evaluation of the patient’s satisfaction in transtibial prosthetics using fusion of variables. The questionnaire technique is used to validate our assertion. The results were very good, but the weighting coefficients were determined empirically.
Comments
No have any comment !