研究目的
To develop a non-contact respiratory monitoring method using a web camera for radiotherapy, addressing the issue of respiratory movement in thoracoabdominal tumors.
研究成果
The proposed web-camera-based method demonstrated strong correlation and equivalence with the conventional Abches device in monitoring respiratory waveforms across different breathing rates. It offers a non-contact, simple, and potentially cost-effective alternative for respiratory monitoring in radiotherapy, with feasibility for clinical use pending further refinements and validations.
研究不足
The study used a simulator rather than human subjects, which may not fully replicate clinical conditions. The image processing relies on a fixed threshold for binarization, which might not be robust to variations in lighting or background. Irregular breathing patterns (e.g., baseline drift, amplitude fluctuations) were not tested, and camera positioning effects on reproducibility require further investigation.
1：Experimental Design and Method Selection:
The study designed an experiment to capture thoracic and abdominal motion images using a web camera from the lateral side of a supine resuscianne simulator, simulating human respiration. Simple image-processing techniques were employed to analyze the motion and generate respiratory waveforms.
2：Sample Selection and Data Sources:
A resuscianne simulator (Laerdal Medical Corp.) was used to simulate respiration at four different rates (12, 17, 22, 27 breaths per minute), representing typical adult and elderly breathing patterns.
3：List of Experimental Equipment and Materials:
Web cameras (ucamdla200h, Elecom Co., Ltd.), resuscianne simulator, Abches device (Apex Medical, Inc.), MATLAB software (MathWorks, Inc.), and a generic notebook PC (CPU
4：5 GHz, memory 0 GB) were used. Experimental Procedures and Operational Workflow:
The web camera was placed approximately 30 cm from the lateral side of the simulator to capture motion images at 30 fps with 640x480 resolution. Images were processed in MATLAB to compute differential images, binarize them with a fixed threshold, calculate area changes for motion detection, and normalize to generate respiratory waveforms. A similar process was applied to Abches indicator images for comparison.
5：Data Analysis Methods:
Correlation coefficients and Bland-Altman analysis were used to compare waveforms from the proposed method and Abches, assessing agreement and similarity.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
