Knee Osteoarthritis Exercise Therapy Normative Research Based On Kalman Filtering Theory

Knee Osteoarthritis is the most common chronic and progressive joint disease of the bone. The disease is developing slowly and not identical to the performance of each patient, but it will result in the loss of function of the knee walking. This will bring patients tremendous psychological and physical trauma and decrease the quality of life in the elderly. Safe and low-cost treatment of KOA has a wide range of social demand and broad application prospects.Exercise therapy can be effective in improving joint function and relieving joint pain, which is a recognized and widely used therapies. However, the normative exercise therapy is judge by the doctor’s experience indicators. It demands patients exercising therapy in the hospital through the guidance of doctors while exercising at home is difficult to ensure therapy normalization, which limits the promotion and application of exercise therapy. Therefore, this paper propose filtering algorithm based on human limb motion estimation methods in conducting research exercise with therapy normative evaluation of new ideas in order to increase patient home exercise therapy training effect. The design steps of the thesis are following:(1) A acceleration signal acquisition and processing. The acceleration signal is collected by a designed signal acquisition system and transfer it to a PC for processing and analyzing.(2) lower limb movement modeling. The paper introduces to describe the acceleration signal of lower limb movement. It makes stress analysis in the straight leg raise during exercise therapy and the establishment of state-space models. The state equation of models reflect the straight leg raising motion rules and the output equation of models reflects the relationship between the acceleration signal and the state variables.(3) The filtering and state estimation. The principles and characteristics and implementation steps of Kalman filter and extended Kalman filter and Unscented Kalman filter are described. collecting Straight leg raising movement data is preprocessing and human limb motion model is estimated by respectively EKF algorithm and UKF algorithm, finally discuss the filtering effect of these two filters.the results show that UKF filtering effect is superior to EKF.(4) Normalization evaluation and monitoring. Judgment rules are determined via the analysis of the state variables to normative influence degree of performance indicators, and then carry out normative judgment. Finally the straight leg raising normative monitoring software based on Android operating system is designed so that normative results can be transformed to mobile phones and in its displaying to has realized the standardization of monitoring.The experimental results show that these two kinds methodes of lower limb motion state estimation based on EKF and UKF can objectively and quantitatively evaluate knee joint exercise therapy normalization, and the monitoring of exercise therapy normative is realized by the straight leg raising normative monitoring software.