基于CT影像组学结合临床特征鉴别肺结核与非结核分枝杆菌肺病的研究

doi:10.19982/j.issn.1000-6621.20230337

摘要/Abstract

摘要：

目的: 探讨基于CT影像组学结合临床特征在鉴别肺结核(pulmonary tuberculosis, PTB)与非结核分枝杆菌肺病(nontuberculous mycobacteria pulmonary disease, NTM-PD)中的价值。方法: 对2019年1月1日至2023年3月31日河南省新乡医学院第一附属医院收治的经细菌培养证实的NTM-PD患者和PTB患者的临床资料及CT图像进行回顾性分析。根据分枝杆菌培养结果将所有患者分为PTB组(75例)和NTM-PD组(58例)。分析患者的临床特征,并将两组间差异有统计学意义的临床特征用于构建临床模型。以CT图像中空洞性病灶作为研究对象,共200个病灶被纳入研究,然后将其按照7∶3的比例随机分为训练集和测试集。使用逻辑回归分类器构建影像组学模型。结合影像组学特征和临床特征构建联合模型,使用受试者工作特征曲线(ROC)及曲线下面积(AUC)、校准曲线评估模型在训练集和测试集中的诊断效能。结果: 单因素分析显示,PTB组年龄[中位数(四分位数)]为45(26,66)岁,与NTM-PD组年龄[63(54,70)岁]比较差异有统计学意义(Z=-3.184,P<0.001);PTB组BMI (19.95±2.83)与NTM-PD组BMI (18.78±2.59)比较差异有统计学意义(t=2.469,P=0.015);PTB组γ-干扰素释放试验(interferon-γ release assays, IGRA)阳性患者55例(73.33 %),NTM-PD组IGRA阳性患者16例(27.59 %),两者阳性率比较差异有统计学意义(χ²=27.505,P<0.001)。多因素分析显示,年龄(OR=0.969,P=0.004)与IGRA(OR=6.026,P<0.001)均是鉴别PTB与NTM-PD的独立预测因子。临床模型的AUC值在训练集和测试集中分别为0.832(95%CI:0.765~0.899)和0.800(95%CI:0.689~0.911);影像组学模型在训练集和测试集中的AUC值分别为0.974(95%CI:0.952~0.996)和0.939(95%CI:0.877~1.000);联合模型在训练集和测试集中的AUC值分别为0.993(95%CI:0.986~1.000)和0.995(95%CI:0.985~1.000)。结论: 临床特征结合影像组学特征的联合模型是一种无创、方便、快捷的辅助诊断方法,能够快速有效地鉴别PTB与NTM-PD。

关键词: 结核,肺, 分枝杆菌感染, 诊断,鉴别, 体层摄影术,X线计算机, 影像组学

Abstract:

Objective: To explore the value of combining CT-based radiomics with clinical features in distinguishing pulmonary tuberculosis (PTB) from nontuberculous mycobacteria pulmonary disease (NTM-PD). Methods: A retrospective analysis was conducted on clinical data and CT images of NTM-PD and PTB patients confirmed with culture from The First Affiliated Hospital of Xinxiang Medical University from January 1, 2019, to March 31, 2023. Based on the results of bacterial culture, all patients were divided into the PTB group (58 cases) and the NTM-PD group (75 cases). Clinical features were analyzed, and statistically significant features were used to construct a clinical model. CT images were used to study cavitary lesions, with a total of 200 lesions included in the study. The lesions were randomly divided into a training set and a testing set in a 7∶3 ratio. A logistic regression classifier was used to construct a radiomics model. A combined model was built by integrating radiomics features and clinical features. The diagnostic performance of the models in the training and testing sets was evaluated by sensitivity, specificity, accuracy, receiver operating characteristic (ROC) curve, area under the curve (AUC), and calibration curve. Results: Univariate analysis showed that there was statistically significant difference in age between the PTB group (median (quartile) age 45 (26, 66) years) and the NTM-PD group (median (quartile) age 63 (54, 70) years)(Z=-3.184, P<0.001). There was also statistically significant difference in BMI between the PTB group (19.95±2.83) and the NTM-PD group (18.78±2.59)(t=2.469, P=0.015). The proportions of patients with positive interferon-gamma release assays (IGRA) results were significantly different between the PTB group (55 cases, 73.33%) and the NTM-PD group (16 cases, 27.59%)(χ²=27.505, P<0.001). Multivariate analysis showed that age (OR=0.969, P=0.004) and IGRA (OR=6.026, P<0.001) were independent predictive factors for distinguishing PTB from NTM-PD. The AUC values of the clinical model in the training and testing sets were 0.832 (95%CI: 0.765-0.899) and 0.800 (95%CI: 0.689-0.911), respectively. The AUC values of the radiomics model in the training and testing sets were 0.974 (95%CI: 0.952-0.996) and 0.939 (95%CI: 0.877-1.000), respectively. The AUC values of the combined model in the training and testing sets were 0.993 (95%CI: 0.986-1.000) and 0.995 (95%CI: 0.985-1.000), respectively. Conclusion: The combined model incorporating clinical features and radiomics features is a non-invasive, convenient, and rapid diagnostic method that can effectively distinguish PTB from NTM-PD.

Key words: Tuberculosis,pulmonary, Mycobacterium infections, Diagnosis,differential, Tomography,X-ray compute, Radiomics

中图分类号:

R521
R81

姚阳阳, 梁长华, 韩东明, 崔俊伟, 潘犇, 王慧慧, 魏正琦, 甄思雨, 危涵羽. 基于CT影像组学结合临床特征鉴别肺结核与非结核分枝杆菌肺病的研究[J]. 中国防痨杂志, 2024, 46(3): 302-310. doi: 10.19982/j.issn.1000-6621.20230337

Yao Yangyang, Liang Changhua, Han Dongming, Cui Junwei, Pan Ben, Wang Huihui, Wei Zhengqi, Zhen Siyu, Wei Hanyu. Differentiation of pulmonary tuberculosis and nontuberculous mycobacterial pulmonary disease based on computed tomography radiomics combined with clinical features[J]. Chinese Journal of Antituberculosis, 2024, 46(3): 302-310. doi: 10.19982/j.issn.1000-6621.20230337

图/表 12

图1

图2

图3

图4

表1

表2

肺结核与非结核分枝杆菌肺病患者临床特征多因素logistic回归分析

临床特征	β值	s $x ˉ$ 值	Wald χ²值	P值	OR(95%CI)值
年龄	-0.031	0.011	8.165	0.004	0.969(0.949~0.990)
BMI	0.144	0.078	3.447	0.063	1.155(0.992~1.345)
IGRA	1.796	0.412	18.966	<0.001	6.026(2.685~13.525)

表2

图5

表3

图6

图7

图8

图9

参考文献 16

[1]	World Health Organization. Global tuberculosis report 2022. Geneva: World Health Organization, 2022.
[2]	中华医学会结核病学分会. 非结核分枝杆菌病诊断与治疗指南(2020年版). 中华结核和呼吸杂志, 2020, 43(11):918-946. doi:10.3760/cma.j.cn112147-20200508-00570.
[3]	李朝俊, 孙诚鸿, 金文宇. 非结核分枝杆菌肺病与耐多药肺结核的胸部CT征象对照分析. 临床荟萃, 2021, 36(6):530-534. doi:10.3969/j.issn.1004-583X.2021.06.010.
[4]	Wittram C, Weisbrod GL. Mycobacterium avium complex lung disease in immunocompetent patients: radiography-CT correlation. Br J Radiol, 2002, 75(892):340-344. doi:10.1259/bjr.75.892.750340.
[5]	Ma J, Zhou Z, Ren Y, et al. Computerized detection of lung nodules through radiomics. Med Phys, 2017, 44(8):4148-4158. doi:10.1002/mp.12331. pmid: 28494110
[6]	Coroller TP, Agrawal V, Huynh E, et al. Radiomic-Based Pathological Response Prediction from Primary Tumors and Lymph Nodes in NSCLC. J Thorac Oncol, 2017, 12(3):467-476. doi:10.1016/j.jtho.2016.11.2226. pmid: 27903462
[7]	Gopalaswamy R, Shanmugam S, Mondal R, et al. Of tuberculosis and non-tuberculous mycobacterial infections-a comparative analysis of epidemiology,diagnosis and treatment. J Biomed Sci, 2020, 27(1):74. doi:10.1186/s12929-020-00667-6. pmid: 32552732
[8]	Duan H, Han X, Wang Q, et al. Clinical Significance of Nontuberculous Mycobacteria Isolated From Respiratory Specimens in a Chinese Tuberculosis Tertiary Care Center. Sci Rep, 2016, 6:36299. doi:10.1038/srep36299. pmid: 27808247
[9]	Liu CF, Song YM, He WC, et al. Nontuberculous mycobacteria in China: incidence and antimicrobial resistance spectrum from a nationwide survey. Infect Dis Poverty, 2021, 10(1):59. doi:10.1186/s40249-021-00844-1.
[10]	Jia H, Pan L, Wang G, et al. Assessment of Interferon-Gamma Release Assay in Patients with Non-Tuberculous Mycobacteria Pulmonary Disease. Clin Lab, 2019, 65(10):10.7754/Clin.Lab.2019.181247. doi:10.7754/Clin.Lab.2019.181247.
[11]	洪创跃, 李金莉, 赵广录, 等. 2013—2017年深圳市非结核分枝杆菌流行状况分析. 中国防痨杂志, 2019, 41(5):529-533. doi:10.3969/j.issn.1000-6621.2019.05.011.
[12]	Yang C, Luo X, Fan L, et al. Performance of Interferon-Gamma Release Assays in the Diagnosis of Nontuberculous Mycobacterial Diseases-A Retrospective Survey From 2011 to 2019. Front Cell Infect Microbiol, 2021, 10:571230. doi:10.3389/fcimb.2020.571230.
[13]	Kang WL, Wang GR, Wu MY, et al. Interferon-Gamma Release Assay is Not Appropriate for the Diagnosis of Active Tuberculosis in High-Burden Tuberculosis Settings: A Retrospective Multicenter Investigation. Chin Med J (Engl), 2018, 131(3):268-275. doi:10.4103/0366-6999.223860.
[14]	Yan Q, Wang W, Zhao W, et al. Differentiating nontuberculous mycobacterium pulmonary disease from pulmonary tuberculosis through the analysis of the cavity features in CT images using radiomics. BMC Pulm Med, 2022, 22(1):4. doi:10.1186/s12890-021-01766-2. pmid: 34991543
[15]	谢智恩, 黎惠如, 宋敏, 等. 非结核分枝杆菌肺病、活动性肺结核以及耐多药肺结核的CT影像对比分析. 医学影像学杂志, 2020, 30(12):2224-2227.
[16]	Ying C, Li X, Lv S, et al. T-SPOT with CT image analysis based on deep learning for early differential diagnosis of nontuberculous mycobacteria pulmonary disease and pulmonary tuberculosis. Int J Infect Dis, 2022, 125:42-50. doi:10.1016/j.ijid.2022.09.031.

临床特征	PTB组(75例)	NTM-PD组(58例)	统计检验值	P值
性别			χ²=0.950	0.330
男性	55(73.33)	38(65.52)
女性	20(26.67)	20(34.48)
年龄(岁)	45(26,66)	63(54,70)	Z=-3.184	<0.001
BMI	19.95±2.83	18.78±2.59	t=2.469	0.015
吸烟			χ²=1.597	0.206
是	38(50.67)	23(39.66)
否	37(49.33)	35(60.34)
饮酒			χ²=0.010	0.922
是	15(20.00)	12(20.69)
否	60(80.00)	46(79.31)
IGRA			χ²=27.505	<0.001
阳性	55(73.33)	16(27.59)
阴性	20(26.67)	42(72.41)
结核抗体			χ²=1.404	0.236
阳性	28(37.33)	16(27.59)
阴性	47(62.67)	42(72.41)
发热			χ²=3.225	0.073
是	32(42.67)	16(27.59)
否	43(57.33)	42(72.41)
盗汗			χ²=0.966	0.326
是	9(12.00)	4(6.90)
否	66(88.00)	54(93.10)
咳嗽			χ²=3.374	0.066
是	64(85.33)	42(72.41)
否	11(14.67)	16(27.59)
咳痰			χ²=3.105	0.078
是	56(74.67)	35(60.34)
否	19(25.33)	23(39.66)
胸痛			χ²=0.260	0.610
是	7(9.33)	7(12.07)
否	68(90.67)	51(87.93)
乏力			χ²=0.047	0.829
是	10(13.33)	7(12.07)
否	65(86.67)	51(87.93)
体质量减轻			χ²=0.008	0.928
是	16(21.33)	12(20.69)
否	59(78.67)	46(79.31)

模型	AUC(95%CI)值	敏感度(%)	特异度(%)	准确率(%)
训练集
临床模型	0.832(0.765~0.899)	89.04	72.46	80.99
影像组学模型	0.974(0.952~0.996)	91.78	95.65	93.66
联合模型	0.993(0.986~1.000)	100.00	91.30	95.77
测试集
临床模型	0.800(0.689~0.911)	85.29	67.86	77.42
影像组学模型	0.939(0.877~1.000)	97.06	82.14	90.32
联合模型	0.995(0.985~1.000)	94.12	100.00	96.77