Traditional methods of non-contact human respiratory rate measurement usually require complex devices or algorithms. Aiming at this problem, a non-contact respiratory rate measurement method based on only the RGB video information was proposed in this paper. The method consisted of four steps. Firstly, spatial filtering was applied to each frame of the input video. Secondly, a gray compensation algorithm was used to compensate for the gray level change caused by the environmental light. Thirdly, the gray levels of each pixel over time were filtered separately by a low-pass filter. Finally, the region of interest was determined based on the filtering results, and the respiration rate of the human is measured. The physical measurement experiments were designed, and the measurement accuracy was compared with that of the biological radar. The error of the proposed method was between − 5.5% and 3% in different detection directions. The results show that the non-contact respiration rate measurement method can effectively measure the human respiration rate.
Objective To develop a Matlab toolbox to improve the efficiency of musculoskeletal kinematics analysis while ensuring the consistency of musculoskeletal kinematics analysis process and results. Methods Adopted the design concept of “Batch processing tedious operation”, based on the Matlab connection OpenSim interface function ensures the consistency of musculoskeletal kinematics analysis process and results, the functional programming was applied to package the five steps for scale, inverse kinematics analysis, residual reduction algorithm, static optimization analysis, and joint reaction analysis of musculoskeletal kinematics analysis as functional functions, and command programming was applied to analyze musculoskeletal movements in large numbers of patients. A toolbox called LLMKA (Lower Limbs Musculoskeletal Kinematics Analysis) was developed. Taking 120 patients with medial knee osteoarthritis as the research object, a clinical researcher was selected using the LLMKA toolbox and OpenSim to test whether the analysis process and results were consistent between the two methods. The researcher used the LLMKA toolbox again to conduct musculoskeletal kinematics analysis in 120 patients to verify whether the use of this toolbox could improve the efficiency of musculoskeletal kinematics analysis compared with using OpenSim. Results Using the LLMKA toolbox could analyze musculoskeletal kinematics analysis in a large number of patients, and the analysis process and results were consistent with the use of OpenSim. Compared to using OpenSim, musculoskeletal kinematics analysis was completed in 120 patients using the LLMKA toolbox with only 2 operations were needed to enter the patient body mass data, operating steps decreased by 99.19%, total analysis time by 66.84%, and manual participation time by 99.72%, just need 0.079 1 hour (4 minutes and 45 seconds). Conclusion The LLMKA toolbox can complete a large number of musculoskeletal kinematics analysis in patients with one click in a way that is consistent in process and results with using OpenSim, reducing the total time of musculoskeletal kinematics analysis, and liberating clinical researchers from cumbersome steps, making more energy into the clinical significance of musculoskeletal kinematics analysis results.