Dual-center clinical evaluation of sputum smear automatic microscopy system for detecting tuberculosis

doi:10.19871/j.cnki.xfcrbzz.2022.04.013

Abstract

Abstract: Objective To evaluate the mycobacterium tuberculosis detection performance of ASSMS and compare it with that of the traditional manual way. Method 474 and 451 sputum specimens were retrospectively collected from Eighth Affiliated Hospital of Xinjiang Medical University (Center Ⅰ) and Shenzhen Chronic Disease Control Center Shenzhen (Center Ⅱ), respectively, and mycobacterium tuberculosis tests were separately conducted by doctors and ASSMS after specimens were made into sputum smears. Results were finally compared and analyzed. Result The detection rate of ASSMS was higher than manual detection on two datasets of 925 cases (Center Ⅰ: P<0.05,Center Ⅱ: P>0.05). The results of center Ⅰshowed that the accuracy, sensitivity and specificity of manual and automatic detection were 91.8% vs 90.5%, 74.4% vs 86.7% and 97.7% vs 91.8%, and the ASSMS finally identified 31 sputum samples (25.8%)that was missed by doctors. While the results from center Ⅱ showed that the accuracy, sensitivity and specificity of manual and automatic detection were 90.7% vs 85.8%, 52.3% vs 68.2% and 94.8% vs 87.7%, and the ASSMS identified 21 sputum samples (47.7%) doctors missed. Conclusion Automatic detection may be functioned as an effective assistant to help doctors reduce missed diagnosis rate of tuberculosis.

Key words: Tuberculosis, Sputum smear, Automated microscopy, Mycobacterium tuberculosis, Comparative study

Zhao Doudou, He Huishan, Xia Li, Zheng Qiuting, Qian Lingjun, Hong Kunlei, Lure Fleming Yuan-ming, Tao Lili, Guo Lin. Dual-center clinical evaluation of sputum smear automatic microscopy system for detecting tuberculosis[J]. Electronic Journal of Emerging Infectious Diseases, 2022, 7(4): 57-60.

References

[1] 中华人民共和国国家卫生和计划生育委员会. 肺结核诊断标准(WS 288-2017)[J/CD]. 新发传染病电子杂志, 2018, 3(1): 59-61.
[2] 卢春容, 房宏霞, 陆普选, 等. WHO 2021年全球结核病报告:全球与中国关键数据分析[J/CD]. 新发传染病电子杂志, 2021, 6(4): 368-372.
[3] WORLD HEALTH ORGANIZATION.Global tuberculosis report 2021[EB/OL]. [2021-10-14].https://www.who.int/publications/digital/global-tuberculosis-report-2021.
[4] 任坦坦, 邓国防, 付亮, 等. 2020 WHO全球结核报告:全球与中国关键数据分析[J]. 新发传染病电子杂志, 2020, 5(4): 280-284.
[5] 孙灵利, 赵平, 郭娇, 等. 发光二极管荧光适配器在基层实验室中检测分枝杆菌的效果评价[J]. 现代预防医学, 2013, 40(16): 3037-3040.
[6] WORLD HEALTH ORGANIZATION.Fluorescent Light-Emitting Diode (LED) Microscopy for Diagnosis of Tuberculosis: Policy Statement[M]. Geneva: World Health Organization, 2011.
[7] ANGEBY K, ALVARADOGALVEZ C, PINEDAGARCIA L, et al.Improved sputum microscopy for a more sensitive diagnosis of pulmonary tuberculosis[J]. Int J Tuberc Lung D, 2000, 4(7): 684-687.
[8] STEINGART KR, NG V, HENRY M, et al.Sputum processing methods to improve the sensitivity of smear microscopy for tuberculosis: a systematic review[J]. Lancet Infect Dis, 2006, 6(10): 664-674.
[9] DIVEKAR A, PANGILINAN C, COETZEE G, et al.Automated detection of tuberculosis on sputum smeared slides using stepwise classification[C]// Proceedings of SPIE-the International Society for Optical Engineering. California, United States, 2012. Washington: SPIE, 2012: 120.
[10] LURE YMF, JAEGER S, ANTANI S, 等. 自动化显微镜检测和数字化胸片诊断系统在肺结核筛查中的应用[J/CD]. 新发传染病电子杂志, 2017, 2(1): 5-9.
[11] HUTLANG R, KRISHNAN S, WHITELAW A, et al.Automated detection of tuberculosis in Ziehl-Neelsen-stained sputum smears using two one-class classifiers[J]. J Microsc, 2010, 237(1): 96-102.
[12] PANICKER RO, KALMADY KS, RAJAN J, et al.Automatic detection of tuberculosis bacilli from microscopic sputum smear images using deep learning methods[J]. Biocybern Biomed Eng, 2018, 38(3): 691-699.
[13] PANICKER RO, SOMAN B, SAINI G, et al.A Review of Automatic Methods Based on Image Processing Techniques for Tuberculosis Detection from Microscopic Sputum Smear Images[J]. J Med Syst, 2016, 40(1): 1-13.
[14] 李强, 杨红国, 邓云峰, 等. 抗酸杆菌自动荧光染色和显微镜扫描技术在结核病诊断中的临床评价[J]. 中国防痨杂志, 2018, 40(10): 1075-1078.
[15] 岳友宏, 李艳霞. 提高抗酸杆菌痰涂片镜检阳性率方法探讨及应用评价[J]. 实用预防医学, 2018, 25(10): 1270-1272.
[16] WINTER A, ABERLE DR, HSU W.External validation and recalibration of the Brock model to predict probability of cancer in pulmonary nodules using NLST data[J]. Thorax, 2019, 74(6): 212413.
[17] XIA H, SONG YY, ZHAO B, et al.Multicentre evaluation of Ziehl-Neelsen and light-emitting diode fluorescence microscopy in China[J]. Int J Tuberc Lung D, 2013, 17(1): 107-112.