Tracking and compensation of respiration pattern by an automatic compensation system

Purpose This study aimed to determine the feasibility of using an ultrasound image tracking algorithm (UITA) combined with a respiration compensating system (RCS) to track and compensate the respiration pattern of the diaphragm in real time. Methods Diaphragm motions and various respiration patterns were tracked and captured in volunteers using our previously developed UITA (Kuo et al., J Xray Sci Technol, 2016:875). A diaphragm phantom was placed on a respiration simulation system (RSS) that received signals with different respiration patterns to simulate actual human respiration signals. The RSS was mounted on the RCS, which is 180 cm long and driven by inputting a compensating signal to a linear actuator underneath with and without using a phase‐lead compensator (PLC) (Chuang et al., J Xray Sci Technol, 2015:503). The target displacement was calculated automatically by the UITA and compensated by the RCS. The phantom displacements were observed using a fluoroscopic imaging system on the linear accelerator at the Department of Radiation Oncology, Taipei Medical University Hospital, and the results were also compared with the displacements measured by the UITA and the RSS for correlation verification. In addition, the compensating effect was analyzed after activating the RCS. Results The experimental results indicate a significant correlation between the UITA‐calculated and actual displacements, with a correlation coefficient of up to 91% for the simulated respiration patterns. After activating the RCS, the obtained compensating effect was more than 65%, and even up to 85% if a PLC was used. Moreover, the compensation of 10 extreme respiration patterns of diaphragm was improved significantly through the use of a PLC, with a peak compensating rate of 88.92% being achieved. Finally, compensation effects ranging from 52% to 74% were obtained in 10 human volunteers. Conclusions This study combined ultrasound imaging tracking technology with the RCS to offset the respiration‐induced diaphragm displacement and compensate the various respiration patterns, even including those with baseline‐shift phenomenon in real time with the aid of a noninvasive ultrasound imaging system.