Objective: To evaluate the effectiveness of different laboratory methods for the supplementary diagnosis of pulmonary tuberculosis（PTB）and provide reference data for the early diagnosis of PTB. Methods: A total of 298 suspected PTB patients, who were diagnosed and treated in the outpatient department of Shanghai Tongren Hospital from January 2016 to December 2018, were divided into 3 groups: active PTB (138 cases)，inactive PTB (43 cases) and non-PTB (117 cases) group. Sputum acid-fast staining, MGIT liquid culture system and Xpert MTB/RIF test were performed to detect the sputum specimens. The sensitivity and specificity were compared by Chi-square test. Results: The three methods showed certain significance for distinguishing active PTB, inactive PTB combined with non-PTB (χ 2 values were 89.08, 138.94 and 137.12 respectively, all P＜0.01). There was no significant difference for the positive rate of the three methods between inactive PTB and non-PTB. The sensitivities of acid-fast staining, MGIT liquid culture, Xpert MTB/RIF test and the combination of three methods in the diagnosis of active PTB were 45.7% (63/138), 63.8% (88/138), 65.4% (87/133) and 78.2% (104/133) respectively. The sensitivities of MGIT culture and Xpert MTB/RIF test were significantly higher than that of acid-fast staining (χ 2 was 35.79 and 11.26 respectively，all P＜0.01). There was no significant difference for the sensitivities between MGIT liquid culture and Xpert MTB/RIF test（χ 2 was 29.87, P＞0.05) . The sensitivity of combined detection was higher than that of single detection（χ 2 was 30.84, 64.62, 70.14, respectively, all P＜0.01）. The diagnostic specificities of the three methods and their combination were 99.1%（116/117), 98.3%（115/116）, 99.1%（113/114) and 97.3%（110/113）respectively. There was no significant difference for the specificities of the three methods. Conclusion High sensitivities of MGIT liquid culture and Xpert MTB/RIF test were shown in PTB diagnosis. Combined detection of the three methods may improve the sensitivity of detection.

OBJECTIVE：To investigate the internal mechanism of Schisandra sphenanthera and Schisandra chinensis in determining quality by color （“color discrimination grading ”）of medicinal materials ，and to construct a qualitative identification model based on color quantization value. METHODS ：HPLC method was used to determine the contents of 6 active components from 39 batches of samples. The colorimeter was used to determine 3-color spatial value [lightness value （ΔL＊），red-green value （Δa＊），yellow-blue value （Δb＊）]. SPSS 24.0 statistical software was used to analyze the correlation between the contents of 6 active components and 3-color spatial values. Principal component analysis （PCA）was performed by using SIMCA-P 14.1 software. RESULTS：The linear range of schizandrol A ，schizandrol B ，schisandrin A ，schisandrin B ，schisandrin C ，schisantherin A were 0.204 8-2.560 0，0.049 3-0.616 3，0.098 4- 1.230 0，0.046 3-0.578 8，0.010 6-0.132 0，0.100 0-1.500 0 μg（r＞0.999 0）；RSDs of precision ，stability（12 h）and repeatability tests were all less than 3％. The recoveries were 98.14％-101.53％（RSD＝1.08％， n＝6），97.16％-101.05％（RSD＝1.54％，n＝6），98.29％-101.41％（RSD＝1.29％，n＝6），97.17％-100.36％（RSD＝1.20％，n＝ 6），97.32％-102.43％（RSD＝1.77％，n＝6）and 98.02％-100.40％（RSD＝0.84％，n＝6），respectively. Among 39 batches of components were 3.25-7.39，0.96-1.98，0.46-4.74，1.62-2.60， 0.06-0.58，0.48-6.11 mg/g，respectively. Average S. chinensis was － 80.79-－ 70.54， average Δ a ＊ was qq.com # 通 2.54-5.34，average Δb＊ was 5.20-12.83，average ΔE＊ was 71.13-81.23；average ΔL＊ of S. sphe nanthera was －75.90- －69.16，average Δa＊ was 3.77-7.82，average Δb＊ was 8.59-17.23，average ΔE＊ was 69.99-77.92. The results of relationship analysis showed that the contents of schizandrol A ，schizandrol B ，schisandrin A ，schisandrin B and schisantherin A were significantly correlated with ΔL＊，Δa＊，ΔE＊（P＜0.01），with no significant correlation with Δb＊（P＞0.05）. There was a negative correlation of the content of schisandrin C with ΔL＊ and Δa＊（P＜0.05），and there was no significant correlation with Δb＊ and ΔE＊ （P＞0.05）. Results of PCA showed that accumulative variance contribution rate of primary 2 main components was 89.8％，and S. sphenanthera and S. chinensis could be identified significantly. CONCLUSIONS ：The content of schizandrol A in S. chinensis is high relatively ，and content of schisantherin A in S. sphenanthera is high relatively. Schizandrol A ，schizandrol B and schisandrin B were not detected in S. sphenanthera . The 3-color spatial value of S. sphenanthera and S. chinensis are different ，that is ，the brightness of S. chinensis is small and the color is slant black ，while the color of S. sphenanthera is slant red and yellow. The contents of active components of S. sphenanthera and S. chinensis is related to the surface 3-color spatial values ，that is ，the darker the color is ，the weaker the red degree is ，and the higher the contents of schizandrol A ，schizandrol B ，schisandrin B and schisandrin C are ；the brighter the surface color is ，the stronger the red degree is ，and the higher the contents of schisandrin A and schisantherin A are. The established content determination method is precise and stable ，and can be used for the content determination of S. sphenanthera and S. chinensis . The color qualitative identification model can be used for the identification of S. sphenanthera and S. chinensis .