纤维化型间质性肺疾病的定量CT与机器学习分类
本研究回顾性纳入1085例经病理证实的UIP、NSIP和CHP患者,评估定量CT(QCT)特征及基于QCT特征的机器学习模型对间质性肺疾病(ILD)的分类效能,并与深度学习(DL)模型比较。结果显示多数QCT特征可区分三种ILD,QCT-ML模型多分类AUC micro平均值0.910–0.925、二分类AUC 0.880–0.899,均高于DL模型(多分类0.798、二分类0.869);M3较M2在多分类预测中差异有统计学意义(∆AUC 0.015,CI [0.002, 0.029])。作者认为需进一步研究确定QCT-ML、DL或两者联合在ILD分类中的优劣。
要点
- QCT特征可区分病理证实的UIP、NSIP和CHP,69/78(88%)的特征在三种ILD间差异有统计学意义(校正p≤0.05)。
- QCT-ML模型多分类AUC micro平均值0.910–0.925、二分类AUC 0.880–0.899,均高于DL模型(0.798、0.869)。
- M3较M2在多分类预测中AUC提高0.015(CI [0.002, 0.029]),差异有统计学意义。
核心结论
- QCT特征能成功区分病理证实的UIP、NSIP和CHP;基于QCT的ML模型分类性能优于DL模型,但需进一步研究确定QCT-ML、DL或两者联合在ILD分类中是否更优。
研究设计与样本
- 回顾性研究,纳入1085例经病理证实的UIP、NSIP或CHP且接受活检前后胸部CT的患者;QCT特征、人口学资料和肺功能检查(PFT)结果用于训练极端梯度提升(XGBoost)模型(训练/验证集n=911),构建M1(仅QCT特征)、M2(M1+年龄和性别)、M3(M2+PFT结果)三个模型,并开发DL模型;以病理为参考标准。
诊断标准与关键数值
- QCT特征与ILD的关联用Kruskal-Wallis检验评估,78个QCT特征中69个(88%)可区分三种ILD(校正p≤0.05);多分类AUC micro平均值:M1 0.910、M2 0.910、M3 0.925、DL 0.798;多分类AUC macro平均值:M1 0.895、M2 0.893、M3 0.925、DL 0.779;二分类(UIP vs 非UIP)AUC:M1 0.880、M2 0.899、M3 0.898、DL 0.869;M3 vs M2多分类∆AUC 0.015(CI [0.002, 0.029])。
临床应用与阅片要点
- 作者认为QCT-ML模型在区分病理证实的UIP、NSIP和CHP方面表现良好且优于DL模型;作者指出需进一步研究确定QCT-ML、DL或两者联合是否在ILD分类中更优。
局限性
- 原文结论指出需进一步研究(further investigations are warranted)以确定QCT-ML、DL或两者联合在ILD分类中是否更优。
- 原文要点指出需额外研究(additional investigations are needed)以确定两种方法或其联合是否提供更优诊断性能。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
OBJECTIVES: To evaluate quantitative computed tomography (QCT) features and QCT feature-based machine learning (ML) models in classifying interstitial lung diseases (ILDs). To compare QCT-ML and deep learning (DL) models' performance.
METHODS: We retrospectively identified 1085 patients with pathologically proven usual interstitial pneumonitis (UIP), nonspecific interstitial pneumonitis (NSIP), and chronic hypersensitivity pneumonitis (CHP) who underwent peri-biopsy chest CT. Kruskal-Wallis test evaluated QCT feature associations with each ILD. QCT features, patient demographics, and pulmonary function test (PFT) results trained eXtreme Gradient Boosting (training/validation set n = 911) yielding 3 models: M1 = QCT features only; M2 = M1 plus age and sex; M3 = M2 plus PFT results. A DL model was also developed. ML and DL model areas under the receiver operating characteristic curve (AUC) and 95% confidence intervals (CIs) were compared for multiclass (UIP vs. NSIP vs. CHP) and binary (UIP vs. non-UIP) classification performances.
RESULTS: The majority (69/78 [88%]) of QCT features successfully differentiated the 3 ILDs (adjusted p ≤ 0.05). All QCT-ML models achieved higher AUC than the DL model (multiclass AUC micro-averages 0.910, 0.910, 0.925, and 0.798 and macro-averages 0.895, 0.893, 0.925, and 0.779 for M1, M2, M3, and DL respectively; binary AUC 0.880, 0.899, 0.898, and 0.869 for M1, M2, M3, and DL respectively). M3 demonstrated statistically significant better performance compared to M2 (∆AUC: 0.015, CI: [0.002, 0.029]) for multiclass prediction.
CONCLUSIONS: QCT features successfully differentiated pathologically proven UIP, NSIP, and CHP. While QCT-based ML models outperformed a DL model for classifying ILDs, further investigations are warranted to determine if QCT-ML, DL, or a combination will be superior in ILD classification.
KEY POINTS: • Quantitative CT features successfully differentiated pathologically proven UIP, NSIP, and CHP. • Our quantitative CT-based machine learning models demonstrated high performance in classifying UIP, NSIP, and CHP histopathology, outperforming a deep learning model. • While our quantitative CT-based machine learning models performed better than a DL model, additional investigations are needed to determine whether either or a combination of both approaches delivers superior diagnostic performance.
引用
- Koo, Williams, Liu, et al. Quantitative CT and machine learning classification of fibrotic interstitial lung diseases. Eur Radiol. 2022. PMID: 35678861. doi:10.1007/s00330-022-08875-4
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。