基于双能X射线透视成像的肺部肿瘤运动跟踪方法及临床评价

目的:利用双能X射线透视成像技术，通过呼吸周期内高低能运动图像序列对软组织剪影图像配准，提供一种无需金属标记的肺部肿瘤呼吸运动跟踪方法。方法:以肺部肿瘤患者的高低能X射线透视图像为研究对象，通过自动双能减影算法获得软组织减影图像，然后采用本研究提出的一种结合自适应参考图像选择和归一化互信息匹配的肿瘤运动跟踪算法，计算肿瘤呼吸运动轨迹和运动幅度。采集并分析19例肺癌患者的临床数据，以人工测量结果为参考基准，评价肿瘤运动跟踪算法的准确性。结果:19例病例分析结果显示，运动跟踪算法计算获得的肿瘤呼吸运动轨迹和运动幅度，与人工测量方法获得的结果具有很好的一致性。对大部分病例，头脚方向运动幅度明显大于左右和腹背方向运动幅度，位于肺下半部分肿瘤的运动幅度明显大于位于肺中上部肿瘤。结论:无需金属标记的肿瘤运动跟踪算法，利用双能减影软组织图像，直接对肿瘤进行图像配准，能够较准确地跟踪肺部肿瘤呼吸运动。

Lung tumor motion tracking method and clinical evaluation based on dual energy X ray fluoroscopy

Objective: To provide a breathing motion tracking method for lung tumors without metal markers by registration of soft tissue subtraction images with the high and low energy motion image sequences in the respiratory cycle using dual-energy X-ray fluoroscopy imaging. Methods:Taking the high and low energy X-ray images of lung cancer patients, soft tissue images were obtained by an automatic dual energy subtraction algorithm. Combining adaptive reference image selection and normalized mutual information matching, a tumor motion tracking algorithm was proposed to calculate the breathing motion curve and motion amplitude of lung tumor. The clinical data of 19 patients with lung cancer were collected and analyzed. The accuracy of tumor motion tracking algorithm was evaluated using the manual measurement results as reference benchmark. Results: For all 19 patients, the results of breathing motion curve and motion amplitude calculated by tumor motion tracking algorithm were in good agreement with the results obtained by manual measurement. In most cases, the motion range in superior/inferior direction was greater than that in left/right and anterior/posterior directions, and the motion range of tumors in the lower half of the lung was larger than that in the middle and upper lung. Conclusion: Using dual-energy subtraction soft tissue images, a tumor motion tracking algorithm without metal markers can directly register tumors, and track the respiratory movement of lung tumors accurately.