上海交通大学学报（医学版）

上海交通大学学报（医学版） ›› 2024, Vol. 44 ›› Issue (8): 1030-1036.doi: 10.3969/j.issn.1674-8115.2024.08.012

• 论著 · 临床研究 • 上一篇    

多模态肺结节诊断模型的临床验证及应用价值探索

许万星1,2(), 王琳2, 郭巧梅2, 王薛庆2, 娄加陶1,2,3()   

  1. 1.江苏大学医学院基础医学研究所，镇江 212013
    2.上海交通大学医学院附属第一人民医院检验医学中心，上海 200080
    3.上海交通大学医学院医学技术学院，上海 200025
  • 收稿日期:2024-01-10 接受日期:2024-04-30 出版日期:2024-08-28 发布日期:2024-08-27
  • 通讯作者: 娄加陶 E-mail:wanxing_Xu@163.com;loujiatao@126.com
  • 作者简介:许万星（1997—），女，硕士生；电子信箱： wanxing_Xu@163.com。
  • 基金资助:
    上海市卫健委协同创新集群项目(2019CXJQ03);上海市第一人民医院特色研究项目(CTCCR-2021B06)

Clinical validation and application value exploration of multi-modal pulmonary nodule diagnosis model

XU Wanxing1,2(), WANG Lin2, GUO Qiaomei2, WANG Xueqing2, LOU Jiatao1,2,3()   

  1. 1.Institute of Basic Medical Sciences, School of Medicine, Jiangsu University, Zhenjiang 212013, China
    2.Clinical Laboratory Medicine Center, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200080, China
    3.College of Health Science and Technology, Shanghai Jiao Tong University School of Medicine, Shanghai 200025, China
  • Received:2024-01-10 Accepted:2024-04-30 Online:2024-08-28 Published:2024-08-27
  • Contact: LOU Jiatao E-mail:wanxing_Xu@163.com;loujiatao@126.com
  • Supported by:
    Innovation Group Project of Shanghai Municipal Health Commission(2019CXJQ03);Clinical Research Innovation Plan of Shanghai General Hospital(CTCCR-2021B06)

RichHTML

18

PDF (PC)

192

摘要：

目的·验证采用随机森林算法并基于血清代谢指纹数据、蛋白标志物癌胚抗原（carcinoembryonic antigen，CEA）和Image-AI的多模态肺结节诊断模型（a multi-modal pulmonary nodule diagnosis model combined metabolic fingerprints，protein biomarker CEA and Image-AI via random forest，MPI-RF）的性能，探索其临床应用价值。方法·入组就诊于上海交通大学医学院附属胸科医院且低剂量螺旋CT表现为肺结节的患者289例，根据术后病理结果将其分为恶性结节组（ n=197）和良性结节组（ n=92），收集并比较2组患者的基本信息。使用电化学发光法检测2组患者术前血清CEA水平，使用基质辅助激光解吸电离质谱（matrix-assisted laser desorption/ionization mass spectrometry，MALDI-MS）检测血清代谢指纹图谱，使用CT影像人工智能模型Image-AI计算影像得分。将CEA数据、血清代谢指纹数据和影像得分整合后输入至MPI-RF中，计算每位患者的恶性概率得分。采用受试者操作特征曲线（receiver operator characteristic curve，ROC曲线）、曲线下面积（area under the curve，AUC）评估不同模型的性能并采用DeLong检验进行比较分析，包括MPI-RF在不同类型（实性、纯磨玻璃、混合磨玻璃）和大小（直径<8 mm、直径≥8 mm）的肺结节中的诊断性能，MPI-RF与Mayo Clinic 模型、美国退伍军人管理局（veterans administration，VA）模型、Brock模型的诊断性能比较，以及MPI-RF与肺部影像报告和数据系统（lung imaging reporting and data system，Lung-RADS）在良恶性结节中的诊断性能比较。结果·MPI-RF在肺结节良恶性鉴别中具有良好的诊断性能（AUC=0.887，95% CI 0.848~0.925，灵敏度为81.22%，特异度为83.70%）；其中，MPI-RF对实性结节的AUC为0.877（95% CI 0.820~0.934），混合磨玻璃结节的AUC为0.858（95% CI 0.771~0.946），纯磨玻璃结节的AUC为0.978（95% CI 0.923~1.000）。对于直径<8 mm的结节，MPI-RF的AUC为0.840（95% CI 0.716~0.963）；直径≥8 mm的结节，其AUC为0.891（95% CI 0.849~0.933）。与现有模型对比的结果显示，MPI-RF的诊断性能优于Mayo Clinic模型、VA模型、Brock模型（均 P=0.000）；与Lung-RADS比较，MPI-RF在总样本、不同类型结节中的诊断性能均较优（均 P=0.000）。结论·MPI-RF是性能优良的良恶性肺结节鉴别诊断模型，具有潜在的临床应用价值。

关键词: 机器学习, 肺腺癌, 代谢组学, 多模态模型, 肺结节

Abstract:

Objective ·To verify the performance and explore the clinical application value of a multi-modal pulmonary nodule diagnosis model combined with metabolic fingerprints, protein biomarker CEA and Image-AI via random forest (MPI-RF). Methods ·This study enrolled 289 patients with pulmonary nodules who were admitted to the Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine and were detected by low-dose helical computed tomography (LDCT). The patients were divided into malignant nodule group ( n=197) and benign nodule group ( n=92) based on postoperative pathological results, and the basic information of the two groups was collected and compared. Electrochemiluminescence was used to detect the preoperative serum CEA levels of the patients in the two groups, matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) was used to detect the serum metabolic fingerprints, and the CT image artificial intelligence model Image-AI was used to calculate the image scores. CEA data, serum metabolic fingerprints data and image scores were integrated and input into MPI-RF to calculate the malignant probability score of each patient. The receiver operator characteristic curve (ROC curve) and area under the curve (AUC) were used to evaluate the performance of different models, and the DeLong test was used for comparative analysis, including the diagnostic performance of MPI-RF in different types (solid nodule, pure ground-glass nodule and part-solid nodule) and sizes (diameter<8 mm and diameter≥8 mm) of pulmonary nodules, the diagnostic performance comparison of MPI-RF with Mayo Clinic model, veterans administration (VA) model and Brock model, and the diagnostic performance comparison of MPI-RF with lung imaging reporting and data system (Lung-RADS) in benign and malignant nodules. Results ·MPI-RF had good diagnostic performance in the differentiation of benign and malignant pulmonary nodules (AUC=0.887, 95% CI 0.848?0.925, sensitivity 81.22%, specificity 83.70%). Among them, the AUC of MPI-RF for solid nodules was 0.877 (95% CI 0.820?0.934), for part-solid nodules was 0.858 (95% CI 0.771?0.946), and for pure ground-glass nodules was 0.978 (95% CI 0.923?1.000). The AUC of MPI-RF was 0.840 (95% CI 0.716?0.963) for nodules within 8 mm diameter and 0.891 (95% CI 0.849?0.933) for nodules larger than 8 mm diameter. Compared with the existing models, the diagnostic performance of MPI-RF was better than that of Mayo Clinic model, VA model and Brock model (all P=0.000). Compared with Lung-RADS, MPI-RF had better diagnostic performance in the total samples and different types of nodules (all P=0.000). Conclusion ·MPI-RF is a model for the differential diagnosis of benign and malignant pulmonary nodules with excellent performance, and has potential clinical application value.

Key words: machine learning, lung adenocarcinoma, metabolomics, multi-modal model, pulmonary nodule

中图分类号: