Investigation of an indirect measurement method of ultrasonic bone cutting forces based on strain sensors

Abstract

The measurement model of ultrasonic bone cutting force is an urgent problem to be solved in the application of ultrasonic osteotomy in robot assisted orthopedic surgery, which is of great significance for achieving bone layer perception and cutting force control during the surgical process. This article proposes an indirect measurement method of ultrasonic bone cutting force based on strain. Firstly, according to the axial vibration strain distribution curve of the ultrasonic osteotomy, the optimal measurement position for the strain gauge is selected as the high strain position (with a high signal-to-noise ratio), and the temperature compensation strategy of the dual strain gauge method is used to reduce or eliminate the thermal effects during the strain measurement. Then, the Pearson correlation coefficient between strain and cutting force in different cutting directions is calculated, and there is a strong positive correlation between strain and cutting force. Finally, the least squares method was used to obtain indirect measurement models of ultrasonic bone cutting force based on strain in different cutting directions. The experimental results showed that the waveform of cutting force measured indirectly based on strain was in good agreement with the waveform of cutting force measured directly by the force sensor. The error between the effective value of indirect measurement of cutting force and the effective value of direct measurement of cutting force did not exceed 13%, which verified the feasibility and effectiveness of this method. The indirect measurement models of ultrasonic bone cutting force based on strain will become a cost-effective choice with good accuracy and high sensitivity, providing a theoretical basis for orthopedic robots to perceive and control cutting force, and promoting the clinical application of ultrasonic osteotomy in orthopedic robots.

A lightweight license plate detection algorithm based on deep learning

“…The number of parameters of +GAM is 4.02 M, and the number of parameters of SYOLOv5 is 3.22 M. In contrast, the number of parameters of the model of the proposed algorithm in this paper is 5.07 M, which is 28.2% less than that of the original model. In order to further verify the effectiveness of this paper's algorithm, this paper and the classic lightweight model of license plate recognition for performance comparison, the literature [4][5][6][7][8] is a lightweight network for the improvement of Shuf-fleNetv2, in order to ensure that the comparison algorithm has a unified evaluation standard [9][10][11][12][13][14][15][16][17][18][19], to ensure the performance of the network processing [20][21][22][23][24][25][26][27][28], we will be the above algorithms in the perfor mance comparison on the CCPD dataset, and the results of the comparison are shown in Table 5. Through data comparison we can find that the improved algorithms in this paper have the highest accuracy rate, the smallest number of parameters, and the shortest running time.…”

The high-temperature gas soft measurement method based on the temperature attenuation patterns of flowing argon gas inside a cavity

“…To overcome this limitation, combined with theoretical calculations, indirect measurement methods offer an effective way to obtain temperature information [17] proposed a method to calculate the spatial temperature distribution within a gradient coil system, obtaining steady-state solutions involving the Fourier series for the temperature distribution on the coil. Based on the indirect measurement model, Linwei Wang [18] implemented ultrasonic bone cutting force measurement through strain sensors, and the error did not exceed 13%.…”