《中国医学装备》杂志社

目的：探索四象限铅栅方位、计数和源距的四象限铅栅模体法对单光子发射计算机断层成像(SPECT)固有空间分辨力和线性影响，为SPECT质量控制与性能检测方法提供参考。方法：依据国家卫生行业标准《伽玛照相机、单光子发射断层成像设备(SPECT)质量控制检测规范》(WS523-2019)，将点源放置在距探头铅栅模体表面1.5、1.7、1.9、2.1和2.3 m(1.5～2.3 m)，使用四象限铅栅模体法对北京协和医学院肿瘤医院在用的两台SPECT设备双探头所获取的图像固有空间分辨力和线性进行检测，每个点源位置分别采集安装四象限铅栅模体后的计数(60、100、200、300、500、600和1 000 k)，基于目测判断可分辨的四象限铅栅模体最小象限公式定量计算线源图像的半高宽及目测不同象限线性的畸变程度，检测获取两台SPECT设备双探头的固有空间分辨力和固有空间线性，并分析四象限铅栅不同方位图像可分辨最小象限和线性的一致性。结果：源距固定，当四象限铅栅模体采集计数依次为60 k、100 k、200～300 k、500～1 000 k时，获取两台SPECT设备双探头四象限铅栅不同方位图像，四象限铅栅8个方位目测可分辨的最小象限即固有空间分辨力同时依次为4种情况，即四象限铅栅各象限不可分辨、第四象限(6.125 mm)、第三象限(5.25 mm)、第二象限(4.375 mm)；固有空间线性均为不可分辨、无线性畸变。四象限铅栅模体检测图像采集计数固定，检测源距1.5～2.3 m时，两台SPECT设备双探头固有空间分辨力和线性检测结果相同。结论：四象限铅栅模体法可完成对SPECT设备固有空间分辨力和线性检测。检测源距在1.5～2.3 m内变化不会对检测结果产生影响。检测时应将每帧图像采集计数≥200 k，铅栅摆放方位优化为正面或反面4个位置。

Objective: To explore effect of four-quadrant lead grid phantom method, including orientation, count rate, and source distance, on the intrinsic spatial resolution and linearity of single photon emission computed tomography(SPECT), so as to take a reference for testing methods of quality control and performance of SPECT. Methods: According to the national health industry standard "Specification for testing of quality control in gamma cameras and single photon emission computed tomography(SPECT)"(WS523-2019), a point source was positioned at 1.5, 1.7, 1.9, 2.1 and 2.3 m(range from 1.5 to 2.3 m) away from the surface of the detector of lead grid phantom. The four-quadrant lead grid phantom method was adopted to detect the intrinsic spatial resolution and linearity of the images that were obtained by dual-detector of two used SPECT equipment at Cancer Hospital of Peking Union Medical College. The counts(60, 100, 200, 300, 500, 600 and 1000 k) were collected respectively at each point source after the four-quadrant lead grid phantom was installed. Based on the minimum quadrant formula of resolvable quadrant of the four-quadrant lead grid phantom that was judged by visual estimation, the full width at half maximum(FWHM) of the line source image, and the degree of linear distortion of different quadrants of visual estimation were quantitatively calculated. The intrinsic spatial resolution and spatial linearity of the dual-detector of two SPECT equipment were detected and obtained, and the consistency of the minimum distinguishable quadrant and linearity in images of different orientations of the four-quadrant lead grid was analyzed. Results: After the source distance was fixed, the images of different orientations of four-quadrant lead grid of dual detector of two SPECT equipment were obtained when the collected counts of the four-quadrant lead grid phantom were respectively 60 k, 100 k, 200-300 k, and 500-1000 k. The distinguishable intrinsic spatial resolutions of simultaneously visual estimation on 8 orientations of fourquadrant lead grid were successively indistinguishable, the 4 th quadrant(6.125 mm), the 3 rd quadrant(5.25 mm), and 2 nd quadrant(4.375 mm). The simultaneously intrinsic spatial linearity were successively, indistinguishable, without linear distortion, without linear distortion, and without linear distortion. The intrinsic spatial resolution and the results of linearity detection in dual-detector of two SPECT equipment were same when the collected count of the detected images of four-quadrant lead grid phantom was fixed, and the detected source distance was from 1.5 m to 2.3 m. Conclusion: The four-quadrant lead grid phantom method can complete the detection for intrinsic spatial resolution and linearity of SPECT equipment. The change of source distance of detection within 1.5 and 2.3 m would not produce effect on the detected results. In detection, the collected count of image of each frame should be ≥200 k, and the placed orientation of lead grid should be optimized at 4 positions of top or bottom.