人工智能模型辅助鼻息肉嗜酸性粒细胞计数的研究

doi:10.3969/j.issn.1006-7795.2026.03.010

首都医科大学学报 ›› 2026, Vol. 47 ›› Issue (3): 490-496.doi: 10.3969/j.issn.1006-7795.2026.03.010

人工智能模型辅助鼻息肉嗜酸性粒细胞计数的研究

娄孝婷¹,荣璐璐¹,王远航²,王晴¹,徐孟珂¹,张亮³,张云岗^1*

1.首都医科大学附属北京朝阳医院病理科，北京 100020；2.北京市陈经纶中学嘉铭分校八九班，北京 100029；3.北京怀柔医院病理科，北京 101499

收稿日期:2026-01-20 修回日期:2026-03-29 出版日期:2026-06-21 发布日期:2026-06-26
通讯作者: 张云岗
基金资助:
国家自然科学基金面上项目（82071068）。

Research on establishing an artificial intelligence model to assist in the counting of eosinophils in nasal polyps

Lou Xiaoting¹, Rong Lulu¹, Wang Yuanhang², Wang Qing¹, Xu Mengke¹, Zhang Liang³, Zhang Yungang^1*

1.Department of Pathology, Beijing Chaoyang Hospital, Capital Medical University, Beijing 100020, China; 2.Grade 8 and 9, Jiaming Branch of Beijing Chenjinglun Middle School, Beijing 100029, China; 3. Department of Pathology, Beijing Huairou Hospital, Beijing 101499, China

Received:2026-01-20 Revised:2026-03-29 Online:2026-06-21 Published:2026-06-26
Supported by:
This study was supported by General Program of National Natural Science Foundation of China (82071068).

摘要/Abstract

摘要： 目的旨在建立并验证一种基于深度学习的人工智能(artificial intelligence, AI)模型，以实现对鼻息肉组织切片中嗜酸性粒细胞的自动识别与定量计数，辅助鼻息肉病理诊断。方法回顾性纳入首都医科大学附属北京朝阳医院220例慢性鼻窦炎伴鼻息肉手术标本，构建包含人工计数与AI计数的对比框架。人工计数由2名高级病理医师及2名初学者完成，在传统显微镜高倍镜下选取4个热点区域进行计数；为尽可能实现视野匹配，AI计数在数字化全视野图像上自动识别4个嗜酸性粒细胞密度最高的热点区域，并计算其平均值。通过皮尔逊相关系数与组内相关系数系统评估方法间及观察者间的一致性。结果 AI模型展现出优异的识别性能，其计数结果与人工计数呈高度正相关（r=0.925, P<0.001），且一致性较好（ICC=0.892, 95% CI: 0.859~0.917）。与之对比，4名计数人员结果的一致性仅为中等水平（ICC=0.750）。结论本研究所构建的AI模型能够实现鼻息肉嗜酸性粒细胞的高精度、高一致性定量分析，显著降低了传统人工方法固有的观察者间变异。虽然P值＜0.001仅反应统计学相关性，其临床意义仍需结合随访数据进一步验证。该模型为嗜酸性粒细胞性鼻息肉的标准化病理诊断提供了客观、高效的技术支持，对推动该疾病的精准分型具有重要价值。

关键词: 慢性鼻窦炎伴息肉, 人工智能, 嗜酸性粒细胞计数, 深度学习, 全视野数字图像, 病理诊断

Abstract: Objective To establish and validate an artificial intelligence (AI) model based on deep learning to automatically identify and quantitatively count eosinophils in nasal polyp tissue sections, thereby assisting in the pathological diagnosis of nasal polyps. Methods A total of 220 surgical specimens from patients with chronic rhinosinusitis with nasal polyps in Beijing Chaoyang Hospital, Capital Medical University were retrospectively included to construct a comparative framework between manual and AI counting. Manual counting were performed by two senior pathologists and two beginners, who selected four hotspots under high magnification of a traditional microscope for counting. To achieve as much field-of-view matching as possible, the AI counting automatically identified the four hotspots with the highest eosinophil density on digital full-field images and calculated their average. The consistency between methods and among observers was systematically evaluated by using Pearson's correlation coefficient and intraclass correlation coefficient. Results The AI model demonstrated excellent recognition performance, with its counting results highly positively correlated with manual counting (r=0.925,P<0.001), and excellent consistency (ICC = 0.892, 95% CI: 0.859-0.917). In contrast, the consistency between the four counters was only moderate (ICC = 0.750). Conclusion The AI model developed in this study can achieve high-precision and high-consistency quantitative analysis of eosinophils in nasal polyps, significantly reducing the inter-observer variation inherent in traditional manual methods. This model provides objective and efficient technical support for the standardized pathological diagnosis of eosinophilic nasal polyps, and it is of great value in promoting the precise classification of this disease.

Key words: chronic rhinosinusitis with nasal polyps, artificial intelligence, eosinophil counting, deep learning, full-field digital images, pathological diagnosis

中图分类号:

R765.41

娄孝婷, 荣璐璐, 王远航, 王晴, 徐孟珂, 张亮, 张云岗. 人工智能模型辅助鼻息肉嗜酸性粒细胞计数的研究[J]. 首都医科大学学报, 2026, 47(3): 490-496.

Lou Xiaoting, Rong Lulu, Wang Yuanhang, Wang Qing, Xu Mengke, Zhang Liang, Zhang Yungang. Research on establishing an artificial intelligence model to assist in the counting of eosinophils in nasal polyps[J]. Journal of Capital Medical University, 2026, 47(3): 490-496.

参考文献

［1］Matsuwaki Y, Ookushi T, Asaka D, et al. Chronic rhinosinusitis: risk factors for the recurrence of chronic rhinosinusitis based on 5-year follow-up after endoscopic sinus surgery［J］. Int Arch Allergy Immunol, 2008, 146(Suppl 1): 77-81.
［2］陈靖媛, 陈雯仪, 罗新, 等. 人工智能辅助慢性鼻窦炎嗅觉障碍预测［J］. 临床耳鼻咽喉头颈外科杂志, 2023, 37(11): 871-877, 885.
［3］Wang K H, Ren Y, Ma L, et al. Explainable artificial intelligence for predicting inflammatory signatures and spatial molecular heterogeneity from nasal polyp histology［J］. J Allergy Clin Immunol, 2025, 156(6): 1573-1587.
［4］Cao P P, Li H B, Wang B F, et al. Distinct immunopathologic characteristics of various types of chronic rhinosinusitis in adult Chinese［J］. J Allergy Clin Immunol, 2009, 124(3):478-484, 484.e1-2.
［5］Lou H F, Meng Y F, Piao Y S, et al. Cellular phenotyping of chronic rhinosinusitis with nasal polyps［J］. Rhinology, 2016, 54(2): 150-159.
［6］Hirsch A G, Stewart W F, Sundaresan A S. Nasal and sinus symptoms and chronic rhinosinusitis in apopulation-based sample［J］. Allergy, 2017, 72(2): 274-281.
［7］Wei B, Liu F, Zhang J, et al. Multivariate analysis of inflammatory endotypes in recurrent nasal polyposis in a Chinese population［J］. Rhinology, 2018, 56(3): 216-226.
［8］Wu Q W, Chen J N, Ren Y, et al. Artificial intelligence for cellular phenotyping diagnosis of nasal polyps by whole-slide imaging［J］. EBioMedicine, 2021, 66: 103336.
［9］Vlaminck S, Vauterin T, Hellings P W, et al. The importance of local eosinophilia in the surgical outcome of chronic rhinosinusitis: a 3-year prospective observational study［J］. Am J Rhinol Allergy, 2014, 28(3): 260-264.
［10］Lou H F, Meng Y F, Piao Y S, et al. Predictive significance of tissue eosinophilia for nasal polyp recurrence in the Chinese population［J］. Am J Rhinol Allergy, 2015, 29(5): 350-356.
［11］McHugh T, Snidvongs K, Xie M, et al. High tissue eosinophilia as a marker to predict recurrence for eosinophilic chronic rhinosinusitis: a systematic review and meta-analysis［J］. Int Forum Allergy Rhinol, 2018, 8(12): 1421-1429.
［12］Yang S K, Kim J W, Won T B, et al. Differences in clinical and immunological characteristics according to the various criteria for tissue eosinophilia in chronic rhinosinusitis with nasal polyps［J］. Clin Exp Otorhinolaryngol, 2023, 16(4): 359-368.
［13］Czyzewski T, Daniel N, Rochman M, et al. Machine learning approach for biopsy-based identification of eosinophilic esophagitis reveals importance of global features［J］. IEEE Open J Eng Med Biol, 2021, 2: 218-223.
［14］吴庆武, 孔维封, 袁联雄, 等. 基于病理全玻片数字化图像的人工智能鼻息肉分型及JESREC诊断标准的对比研究［J］. 中华耳鼻咽喉头颈外科杂志, 2022, 57(2): 136-141.
［15］吕威. 人工智能在慢性鼻窦炎精准诊疗中的研究进展［J］. 中国耳鼻咽喉颅底外科杂志, 2025, 31(3): 1-6.
［16］Long E P, Chen J J, Wu X H, et al. Artificial intelligence manages congenital cataract with individualized prediction and telehealth computing［J］. NPJ Digit Med, 2020, 3: 112.
［17］邱实, 宋鹏飞, 常志红, 等. 基于人工智能的细胞形态学分析在白细胞分类中的临床应用［J］. 实用临床医药杂志, 2023, 27(23): 1-5, 11.
［18］Vande Casteele N, Leighton J A, Pasha S F, et al. Utilizing deep learning to analyze whole slide images of colonic biopsies for associations between eosinophil density and clinicopathologic features in active ulcerative colitis［J］. Inflamm Bowel Dis, 2022, 28(4): 539-546.
［19］Larey A, Aknin E, Daniel N, et al. Harnessing artificial intelligence to infer novel spatial biomarkers for the diagnosis of eosinophilic esophagitis［J］. Front Med, 2022, 9: 950728.［20］Ricaurte Archila L, Smith L, Sihvo H K, et al. Performance of an artificial intelligence model for recognition and quantitation of histologic features of eosinophilic esophagitis on biopsy samples［J］. Mod Pathol, 2023, 36(10): 100285.
［21］Vanstapel A, Vanuytsel T, De Hertogh G. Eosinophilic peak counts in eosinophilic esophagitis: a retrospective study［J］. Acta Gastroenterol Belg, 2019, 82(2): 243-250.
［22］Lhermitte B, Egele C, Weingertner N, et al. Adequately defining tumor cell proportion in tissue samples for molecular testing improves interobserver reproducibility of its assessment［J］. Virchows Arch, 2017, 470(1): 21-27.
［23］Liljequist D, Elfving B, Skavberg Roaldsen K. Intraclass correlation-a discussion and demonstration of basic features［J］. PLoS One, 2019, 14(7): e0219854.

[1]	梁正, 刘揆亮, 宁婷婷, 李鹏, 张澍田, 隗永秋. 人工智能在胆管癌诊疗中的应用[J]. 首都医科大学学报, 2026, 47(1): 190-197.
[2]	王永杰, 崔利宾, 袁鑫, 路茜, 陈学明, 刘亮. 人工智能辅助诊断骨质疏松性椎体压缩骨折的效能分析[J]. 首都医科大学学报, 2025, 46(5): 777-783.
[3]	刘彦怡, 张岩岩, 李宏军. 影像组学联合多组学在肝细胞癌的研究进展[J]. 首都医科大学学报, 2025, 46(5): 934-939.
[4]	王拥军, 刘涛, 刘子阳, 熊云云, 荆京, 谢雪微, 李子孝. 人工智能在脑血管病领域中的应用[J]. 首都医科大学学报, 2025, 46(2): 177-183.
[5]	屈俊达, 杨敏福, 李春林, 孙立伟, 高赫, 张旭. 心外膜脂肪影像分割量化方法及其临床应用的研究进展[J]. 首都医科大学学报, 2025, 46(1): 99-105.
[6]	裴友娟, 阮祥燕, 鞠蕊, 王月姣, 蒋玲玲, 马郡. 人工智能在生殖内分泌领域中的应用[J]. 首都医科大学学报, 2024, 45(4): 566-572.
[7]	罗艺, 秦齐, 王沛豪, 郭康顺, 张鑫平, 崔永. 人工智能结合光谱诊断技术在肺癌组织鉴别领域的应用与研究进展[J]. 首都医科大学学报, 2023, 44(6): 1095-1102.

人工智能模型辅助鼻息肉嗜酸性粒细胞计数的研究

Research on establishing an artificial intelligence model to assist in the counting of eosinophils in nasal polyps

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

本文评价