[1]杜海莲,田春燕,赵 利,等.人工智能与免疫荧光染色结合技术对支气管肺泡灌洗液病原学诊断的快速现场评价应用研究[J].现代检验医学杂志,2026,41(01):170-174+184.[doi:10.3969/j.issn.1671-7414.2026.01.033]
　DU Hailian,TIAN Chunyan,ZHAO Li,et al.Application Study of Artificial Intelligence-Integrated Immunofluorescence Staining Technology for Rapid On-Site Evaluation of Pathogen Diagnosis in Bronchoalveolar Lavage Fluid[J].Journal of Modern Laboratory Medicine,2026,41(01):170-174+184.[doi:10.3969/j.issn.1671-7414.2026.01.033]
点击复制

人工智能与免疫荧光染色结合技术对支气管肺泡灌洗液病原学诊断的快速现场评价应用研究()

《现代检验医学杂志》[ISSN:/CN:]

卷:
第41卷
期数:
2026年01期
页码:
170-174+184
栏目:
论著
出版日期:
2026-01-15

文章信息/Info

Title:
Application Study of Artificial Intelligence-Integrated Immunofluorescence Staining Technology for Rapid On-Site Evaluation of Pathogen Diagnosis in Bronchoalveolar Lavage Fluid
文章编号:
1671-7414（2026）01-170-06
作者:
杜海莲1，田春燕1，赵 利1，张海津1，何明鸿1，袁相恋2，刘 菲1
1.潍坊市益都中心医院呼吸科，山东潍坊 262500;2.江苏诺鬲生物科技有限公司，江苏泰州 225300
Author(s):
DU Hailian1，TIAN Chunyan1，ZHAO Li1，ZHANG Haijin1，HE Minghong1，YUAN Xianglian2，LIU Fei1
1.Department of Respiratory, Yidu Central Hospital of Weifang, Shandong Weifang 262500, China;2.Jiangsu Nuoge Biotechnology Co. Ltd, Jiangsu Taizhou 225300, China
关键词:
快速现场评价; 人工智能; 免疫荧光; 支气管肺泡灌洗液
分类号:
R392-33；R446.19
DOI:
10.3969/j.issn.1671-7414.2026.01.033
文献标志码:
A
摘要:
目的探讨人工智能与免疫荧光染色结合的人工智能免疫荧光快速现场评价（AI-IF-ROSE）在肺部下呼吸道感染中支气管肺泡灌洗液（BALF）病原学快速检测中的临床价值。方法?选取2021年4月~2024年3月在潍坊市益都中心医院呼吸科就诊的448例疑似肺部感染患者为研究对象，患者的BALF标本分别采用AI-IF-ROSE染色法和常规快速现场评价（ROSE）染色法进行病原微生物的检测。以培养法检测结果为金标准，绘制受试者工作特征(ROC)曲线并计算曲线下面积（AUC），分析两种ROSE技术对于BALF病原学的快速诊断价值。结果AI-IF-ROSE染色法的镜下真菌、细菌和结核分枝杆菌形态典型，易于观察辨别。AI-IF-ROSE染色法对于真菌感染、细菌感染和结核分枝杆菌感染的阳性率分别为14.73%、27.00%和10.94%，均高于常规ROSE染色法的阳性率（10.04%、19.87%、7.14%），差异具有统计学意义（χ2=4.535、6.369、3.923，均P<0.05）。培养法检出真菌感染68例，细菌感染122例，结核分枝杆菌感染50例。以培养法结果为诊断标准，ROC曲线分析显示，AI-IF-ROSE组检测真菌、细菌和结核分枝杆菌感染的AUC分别为0.985(95%CI：0.962~1.000)、0.996（95%CI：0.987~1.000）、0.979（95%CI：0.947~1.000），高于常规ROSE组检测的0.831（95%CI：0.761~0.901）、0.865（95%CI：0.816~0.913）、0.820（95%CI：0.737~0.903），差异具有统计学意义（均P<0.001）。结论AI-IF-ROSE技术能够对BALF的微生物做快速的区分和分类，在肺部下呼吸道感染的早期快速筛查和后续精准治疗中具有较高的临床优势和应用价值。
Abstract:
Objective To explore the clinical value of the artificial intelligence immunofluorescence rapid on-site evaluation (AI-IF-ROSE) method, which combines artificial intelligence with immunofluorescence staining technology, for the rapid detection of pathogens in bronchoalveolar lavage fluid (BALF) in lower respiratory tract infections of the lung. Methods A total of 448 pa-tients presenting with suspected lung infection at the respiratory department of Yidu Central Hospital of Weifang from April 2021 to March 2024 were included in this investigation. Patient alveolar lavage fluid samples were tested for pathogenic microorgan-isms by AI-IF-ROSE staining and routine ROSE staining. Using the results of culture method as the gold standard, the receiver operating characteristic (ROC) curve was drawn and the area under the curve (AUC) was calculated to analyze the rapid diagnos-tic value of the two ROSE techniques for the etiology of bronchoalveolar lavage fluid. Results The AI-IF-ROSE staining method demonstrated typical morphology of microscopic fungi, bacteria and Mycobacterium tuberculosis, making it easy to observe and distinguish. The positive detection rates for fungi, bacteria, and Mycobacterium tuberculosis were 14.73%, 27.00% and 10.94%, respectively. These rates were higher than those achieved with the routine ROSE staining method (10.04%, 19.87% and 7.14%, respectively), and the differences were statistically significant (χ2=4.535, 6.369, 3.923, all P<0.05). Culture methods detected 68 cases of fungal infection, 122 cases of bacterial infection, and 50 cases of Mycobacterium tuberculosis infection. Using the results of culture methods as the diagnostic standard, ROC curve analysis showed that the AUC for detecting fungal infections in the AI-IF-ROSE group was 0.985 (95%CI: 0.962~1.000), for bacterial infections was 0.996 (95%CI: 0.987~1.000), and for Mycobacterium tuberculosis infections was 0.979 (95%CI: 0.947~1.000)，these values were higher than those detected by the conventional ROSE group, which were 0.831 (95%CI: 0.761~0.901), 0.865 (95%CI: 0.816~0.913) and 0.820 (95%CI: 0.737~0.903), respec-tively. The differences were statistically significant (all P<0.001). Conclusions AI-IF-ROSE technology can quickly distinguish and classify the microorganisms of bronchoalveolar lavage fluid, and has high clinical advantages and application value in the early rapid screening of pulmonary lower respiratory tract infection and subsequent precision treatment.

参考文献/References:

[1] WENG X W, ZHI L J, AN X, et al. Diagnostic value of rapid on-site evaluation in interventional pulmonology:a protocol for systematic review and meta analysis[J]. Medicine, 2020, 99(29): e21168.
[2] 夏婷婷,左春磊,史家欣,等.快速现场评价在呼吸系统疾病中的应用进展[J].医学研究与战创伤救治,2023,36(1):86-89. XIA T T, ZUO C L, SHI J X, et al. Application progress of rapid on-site evaluation in respiratory diseases[J]. Journal of Medical Research & Combat Trauma Care, 2023, 36(1): 86-89.
[3] AGARWAL P, TOI P C, SUBRAMANIAM H, et al. Prospective comparison of cytological specimen ade-quacy assessment by different rapid staining techniques for rapid on-site evaluation in fine needle aspiration cytology and their cost-effectiveness[J]. Diagnostic Cy-topathology, 2019, 47(5): 469-474.
[4] ZALOUDEK C J , HENDRICKSON M R, SOSLOW R A. Mesenchymal tumors of the uterus; in Kurman RJ, Hedrick Ellenson L, Ronnet B M(eds): Blaustein’s Pa-thology of the Female Genital Tract [M]. 6th Ed. New York:springer, 2011: 455-530.
[5] BENNETT J A, BRAGA A C, PINTO A, et al. Uterine PEComas: a morphologic, immunohistochemical, and molecular analysis of 32 tumors[J]. American Journal of Surgical Pathology, 2018, 42(10): 1370-1383.
[6] 陈丽雅,王凤翔,朱方超,等.免疫荧光染色在幽门螺杆菌检测中的应用价值[J].中国现代医生,2022, 60(21):42-45. CHEN L Y, WANG F X, ZHU F C, et al. Application val-ue of immunofluorescence staining in Helicobacter pylori detection[J]. China Modern Doctor, 2022, 60(21): 42-45.
[7] 吴殿水,王新芳,朱妍妍,等.免疫磁珠荧光染色法在痰液结核分枝杆菌检测中的应用[J].现代检验医学杂志,2023,38(2):96-101. WU D S, WANG X F, ZHU Y Y, et al. Application of immunomagnetic bead fluorescence staining in detection of mycobacterium tuberculosis in sputum[J]. Journal of Modern Laboratory Medicine, 2023, 38(2): 96-101.
[8] 王艳平,郑鹏,田敏,等.痰涂片荧光染色法在老年患者呼吸道真菌感染中的应用[J].检验医学与临床, 2023,20(4):550-553. WANG Y P, ZHENG P, TIAN M, et al. Using ently stained sputum smear in respiratory fungal infections in elderly patients[J]. Laboratory Medicine and Clinic, 2023, 20(4): 550-553.
[9] 陈鸣,崔巍,陈瑜,等.“检验医学”遇上“人工智能”[J],国际检验医学杂志,2020,41(5):513-517, 531. CHEN M, CUI W, CHEN Y, et al. “Laboratory medicine”meets “artificial intelligence”[J]. International Journal of Laboratory Medicine, 2020, 41(5): 513-517, 531.
[10] ANDRAS I, MAZZONE E, VAN LEEUWEN F W B, et al. Artificial intelligence and robotics: a combination that is changing the operating room[J]. World Journal of Urology, 2020, 38(10): 2359-2366.
[11] MAK K K, PICHIKA M R. Artificial intelligence in drug development: present status and future pros-pects[J]. Drug Discovery Today, 2019, 24(3): 773-780.
[12] PARANJAPE K, SCHINKEL M, HAMMER R D, et al. The value of artificial intelligence in laboratory medi-cine[J]. American Journal of Clinical Pathology, 2021,155(6): 823-831.
[13] 杨华,孙天舒,王瑶,等.人工智能辅助阅片与单纯人工阅片在女性阴道微生态系统形态学诊断中的比对研究[J].现代检验医学杂志,2023,38(1):169-174, 198. YANG H, SUN T S, WANG Y, et al. Comparative study of artificial intelligence-assisted analysis and manual visual analysis in gynecological microbiome diagno-sis[J]. Journal of Modern Laboratory Medicine, 2023, 38(1): 169-174, 198.
[14] 陈洁,李文生,张巍.人工智能辅助系统在宫颈液基细胞学分析中的应用价值研究[J].现代检验医学杂志,2023,38(5):155-159. CHEN J, LI W S, ZHANG W. Study on the value of artificial intelligence-assisted systems in cervical liq-uid-based cytology analysis[J]. Journal of Modern Lab-oratory Medicine, 2023, 38(5): 155-159.
[15] 章鼎,尹飞.人工智能在脊髓疾病中的应用及研究进展[J].中国实验诊断学,2024,28(1):125-127,F0003. ZHANG D, YIN F. Application and research progress in artificial intelligence in spinal cord diseases[J]. Chi-nese Journal of Laboratory Diagnosis, 2024, 28(1):125-126,F0003.
[16] 董阿静,赵丹萍,李海波.快速现场评价技术在重症监护室的应用进展[J].临床肺科杂志,2024,29(1):123-127. DONG A J, ZHAO D P, LI H B. Progress in the appli-cation of the rapid field evaluation technology in the intensive care unit[J]. Journal of Clinical Pulmonary Medicine, 2024, 29(1): 123-127.
[17] 覃善君,张朝顺,杨新魁,等.快速现场评价技术在消化道占位性病变诊断中的价值[J].新医学,2020, 51(4):294-298. QIN S J, ZHANG C S, YANG X K, et al. The value of rapid on-site evaluation in the diagnosis of digestive tract space-occupying lesions[J]. Journal of New Medi-cine, 2020, 51(4): 294-298.
[18] 王忠弢,石永生.快速现场评价在现代儿科介入呼吸病学的应用[J].中国实用儿科杂志,2019,34(6):470-475. WANG Z T, SHI Y S. Application of rapid on-site eval-uation in contemporary pediatric interventional respira-tory diseases[J]. Chinese Journal of Practical Pediatrics, 2019, 34(6): 470-475.
[19] NATALI F, CANCELLIERI A, TINELLI C, et al. A trained pulmonologist can reliably assess endosonog-raphy-derived lymph node samples during rapid on-site evaluation[J]. Respiration: International Review of Thoracic Diseases, 2019, 97(6): 540-547.
[20] JAIN D, ALLEN T C, AISNER D L, et al. Rapid on-site evaluation of endobronchial ultrasound-guided transbronchial needle aspirations for the diagnosis of lung cancer: a perspective from members of the pul-monary pathology society[J]. Archives of Pathology &Laboratory Medicine, 2018, 142(2): 253-262.
[21] WANG H S, REN T, WANG X, et al. Rapid on-site evalu-ation of touch imprints of biopsies improves the diagnostic yield of transbronchial biopsy for endoscopically nonvisi-ble malignancy: a retrospective study[J]. Japanese Journal of Clinical Oncology, 2021, 51(4): 622-629.
[22] 王秀丽,王凯飞,解立新. 重症监护病房床旁病原学快速诊断-快速现场微生物学评价[J]. 中华检验医学杂志,2023,46(10):997-1002. WANG X L, WANG K F, XIE L X. Rapid on-site pathogen diagnosis of intensive care unit: microbio-logical rapid on-site evaluation[J]. Chinese Journal of Laboratory Medicine, 2023, 46(10): 997-1002.
[23] HOOVEN T A, POLIN R A. Pneumonia[J]. Seminars in Fetal & Neonatal Medicine, 2017, 22(4): 206-213.
[24] TAO Y, SONG L C, FU H, et al. Application of mi-crobiological rapid on-site evaluation in respiratory intensive care units: a retrospective study[J]. Annals of Translational Medicine, 2022, 10(1): 7.
[25] 何文军,李曼,李涛,等.基于血细胞形态识别的自动检测系统的研发[J].现代检验医学杂志,2019,34(2):104-108. HE W J, LI M, LI T, et al. Study on automatic detection sys-tem base on blood cell morphology recognition[J]. Journal of Modern Laboratory Medicine, 2019, 34(2): 104-108.
[26] 杨嫚.人工智能技术辅助诊断肺小结节的临床应用价值[J].中国现代医药杂志,2024,26(6):75-77. YANG M. Clinical application value of artificial intelligence technology to assist in the diagnosis of small pulmonary nod-ules[J]. Modern Medicine Journal of China, 2024, 26(6): 75-77.
[27] 房宇坤,李玉,王剑,等.微生物学快速现场评估在支气管镜下诊断肺部感染的价值[J].中国当代医药, 2020,27(14):4-7, 29. FANG Y K, LI Y, WANG J, et al. Value of microbiolog-ical rapid on-site evaluation in the diagnosis of pulmo-nary infection under bronchoscope[J]. China Modern Medicine, 2020, 27(14): 4-7, 29.

相似文献/References:

[1]杨 华,孙天舒,王 瑶,等.人工智能辅助阅片与单纯人工阅片在女性阴道微生态系统形态学诊断中的比对研究[J].现代检验医学杂志,2023,38(01):169.[doi:10.3969/j.issn.1671-7414.2023.01.032]
　YANG Hua,SUN Tian-shu,WANG Yao,et al.Comparative Study of Artificial Intelligence-assisted Analysis and Manual Visual Analysis in Gynecological Microbiome Diagnosis[J].Journal of Modern Laboratory Medicine,2023,38(01):169.[doi:10.3969/j.issn.1671-7414.2023.01.032]
[2]冯馨锐,吴韦铷,张晓丽,等.人工智能辅助血细胞形态学检查的机遇与挑战[J].现代检验医学杂志,2025,40(01):189.[doi:10.3969/j.issn.1671-7414.2025.01.036]
　FENG Xinrui,WU Weiru,ZHANG Xiaoli,et al.Opportunities and Challenges of Artificial Intelligence Assisted Blood Cell Morphology Examination[J].Journal of Modern Laboratory Medicine,2025,40(01):189.[doi:10.3969/j.issn.1671-7414.2025.01.036]

备注/Memo

备注/Memo:
基金项目：潍坊市卫健委科研计划项目（WFWSJK-2021-257）。
作者简介：杜海莲（1982-），女，硕士，副主任医师，研究方向：呼吸病学研究，E-mail：duhailian82@163.com。
通讯作者：刘菲（1988-），女，硕士，主治医师，研究方向：呼吸病学和肺恶性肿瘤研究，E-mail：15006619696@163.com。
更新日期/Last Update: 2026-01-15