目的 探讨CT影像组学对肺实性结节性粘液腺癌（PNMA）与肺炎性结节（PIN）鉴别价值。方法 选取2016年1月-2023年7月在我院就诊的51例PNMA和60例PIN患者。采用多层螺旋计算机断层（MSCT）扫描检查，统计比较两组影像特征、临床资料。提取两组影像组学特征，采用Logistic回归分析PNMA发生的临床因素并构建Clinic模型。以支持向量机（SVM）构建最优特征模型（Rad模型）。使用Python3.6基于Softmax策略构建人工神经网络模型（Combine模型）。 结果 PNMA组在病灶大小、IL-6、CRP、CEA及多项影像特征（细小毛刺、分叶、密度不均、空泡、胸膜凹陷、血管集束）方面均高于PIN组（均P<0.05）。分叶征、CRP、空泡征、血管集束征、胸膜凹陷征、均是预测PNMA的独立影响因素（均P<0.05）。融合临床独立因素和影像组学特征的Combine模型表现出最优诊断性能，训练集和验证集AUC分别为0.955和0.940，明显高于单一临床或影像组学模型（Delong检验，P<0.05），且校准度和临床净收益均更优。 结论 基于多层螺旋CT的影像组学分析表明，融合临床独立因素和影像组学特征的Combine模型在PNMA与PIN的鉴别诊断中表现出最优诊断性能，具有较高的准确性和临床应用价值。
Abstract
Objective To explore the value of CT radiomics in differentiating pulmonary nodular mucinous adenocarcinoma (PNMA) from inflammatory pulmonary nodules (PIN). Methods A total of 51 PNMA cases and 60 PIN cases who visited our hospital from January 2016 to July 2023 were selected. Multi-slice spiral computed tomography (MSCT) scans were performed to compare the imaging features and clinical data of two groups. Imaging radiomics features were extracted from both groups, and logistic regression analysis was used to identify clinical factors associated with PNMA and construct a Clinic model. An optimal feature model (Rad model) was built using a support vector machine (SVM). Based on Python 3.6 and the Softmax strategy, an artificial neural network model (Combine model) was constructed. Results PNMA group had higher lesion size, IL-6, CRP, CEA, and multiple imaging features (fine spiculation, lobulation, heterogeneous density, cavitation, pleural retraction, and vascular convergence) compared to the PIN group (all P<0.05). Lobulation, CRP, cavitation, vascular convergence, and pleural retraction were independent predictors of PNMA (all P<0.05). The Combine model, which integrates clinical independent factors and radiomics features, demonstrated the best diagnostic performance, with AUCs of 0.955 and 0.940 in the training and validation sets, respectively, significantly higher than single clinical or radiomics models (Delong test, P<0.05), and showed better calibration and clinical net benefit. Conclusions Radiomics analysis based on multi-slice spiral CT shows that Combine model, which integrates independent clinical factors and radiomics features, demonstrates the best diagnostic performance in distinguishing PNMA from PIN, exhibiting high accuracy, and clinical applicability.
关键词
肺实性结节性粘液腺癌 /
/
/
肺炎性结节 /
/
/
多层螺旋计算机断层扫描 /
/
/
鉴别
Key words
Pulmonary solid nodular mucinous adenocarcinoma /
/
/
Pneumonic nodule /
/
/
Multi-slice spiral computed tomography /
/
/
Differential diagnosis
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Leiter A, Veluswamy R, Wisnivesky J.The global burden of lung cancer: current status and future trends[J]. Nat Rev Clin Oncol, 2023, 10(9):624-639. DOI: 10.1038/s41571-023-00798-3.
[2] 王利平, 段颖佳, 候箭, 等.低剂量胸部CT筛查早期肺癌的临床应用研究[J].现代肿瘤医学, 2021, 29(3):407-409. DOI:10.3969/j.issn.1672-4992.2021.03.009.
Wang LP, Duan YJ, Hou J, et al. Clinical application of low-dose chest CT screening for early-stage lung cancer [J]. Modern Oncology, 2021, 29(3):407-409. DOI: 10.3969/j.issn.1672-4992.2021.03.009.
[3] Travis WD, Brambilla E, Nicholson AG, et al. The 2015 World Health Organization classification of lung tumors: impact of genetic, clinical and radiologic advances since the 2004 classification[J].J Thorac Oncol, 2015, 10(9):1243-1260. DOI:10.1097/JTO.0000000000000630.
[4] Xu J, Erlendsson S, Singh M, et al. PNMA2 forms immunogenic non-enveloped virus-like capsids associated with paraneoplastic neurological syndrome[J]. Cell,187(4):831-845. DOI:10.1016/j.cell. 2024.01.009.
[5] Xiao YD, Lv FJ, Li WJ, et al. Solitary Pulmonary Inflammatory Nodule: CT Features and Pathological Findings[J]. J Inflamm Re, 2021, 6(14):2741-2751. DOI:10.2147/JIR.S304431.
[6] 高院, 宋延峰, 张晓艳, 等.低剂量螺旋CT联合血清肺癌自身抗体在肺癌早期诊断中的临床应用[J]. 疑难病杂志, 2023, 22(1): 36-41. DOI:10.3969/j.issn.1671-6450.2023.01.007.
Gao Y, Song YF, Zhang XY, et al. Clinical application of low-dose spiral CT combined with serum lung cancer autoantibodies in early diagnosis of lung cancer [J]. Journal of Difficult and Complicated Cases, 2023, 22(1): 36-41. DOI: 10.3969/j.issn.1671-6450.2023.01.007.
[7] Reza Z, Arya A, Isa M, et al. Assessment of Barriers and Challenges to Screening, Diagnosis, and Biomarker Testing in Early-Stage Lung Cancers[J].Cancers, 2023, 15(5):1595-1606. DOI:10.3390/cancers1505 1595.
[8] 金利, 俞霞, 张涛, 等.基于CT平扫的影像组学模型鉴别肺炎性肌纤维母细胞瘤和周围型肺癌的研究[J]. 影像诊断与介入放射学, 2021, 30(1): 39-43. DOI:10.3969/j.issn.1005-8001.2021.01.008.
Jin L, Yu X, Zhang T, et al. Study on radiomics model based on CT plain scan to differentiate inflammatory myofibroblastic tumor and peripheral lung cancer [J]. Imaging Diagnosis & Interventional Radiology, 2021, 30(1): 39-43. DOI: 10.3969/j.issn.1005-8001. 2021. 01.008.
[9] Ge L, Wang L, Pei D. Pulmonary mucinous adenocarcinoma: An overview of pathophysiology and advancements in treatment[J].Heliyon,2024, 10(9): 281-288. DOI:10.101 6/j.heliyon.2024.e28881.
[10]Li WJ, Lv FJ, Tan YW, et al. Pulmonary Benign Ground-Glass Nodules: CT Features and Pathological Findings[J].Int J Gen Med, 2021, 14(2): 581-590. DOI:10.2147/IJGM.S298517.
[11]宋兰, 田杜雪, 王金华,等.胸部超低剂量CT应用深度学习重建行肺癌筛查的可行性研究[J].中华放射学杂志, 2022, 56(6):667-672. DOI:10.3760/cma.j.cn112149-20210810-00463.
Song L, Tian DX, Wang JH, et al. Feasibility study of lung cancer screening with ultra-low-dose chest CT using deep learning reconstruction [J]. Chinese Journal of Radiology, 2022, 56(6): 667-672. DOI: 10.3760/cma.j.cn112149-20210810-00463.
[12]Balagurunathan Y, Beers A, Mcnitt-Gray M, et al.Lung Nodule Malignancy Prediction in Sequential CT Scans: Summary of ISBI 2018 Challenge[J].IEEE Trans Med Imaging, 2021, 40(12):3748-3761. DOI:10.1109/TMI.2021.3097665.
[13]亓立勇. CT诊断孤立性肺炎性结节的价值及影像特点观察[J].中国CT和MRI杂志, 2021, 19(11):65-67. DOI:10.3969/j.issn.1672-5131. 2021.11.021.
Qi LY. Value and imaging characteristics of CT in the diagnosis of solitary inflammatory pulmonary nodules [J]. Chinese Journal of CT and MRI, 2021, 19(11):65-67. DOI: 10.3969/j.issn.1672-5131. 2021.11.021.
[14]Yao Y, Xue P, Jian G, et al. Metabolomic differentiation of benign vs malignant pulmonary nodules with high specificity via high-resolution mass spectrometry analysis of patient sera[J]. Nat Commun, 2023, 14(1):2339-2351. DOI: 10.1038/s41467-023-37875-1.
[15]梁拥辉, 郭爱菊, 师新宇, 等.肺孤立性实性结节61例采用CT检查对黏液腺癌与结核球鉴别诊断价值的评估[J].中华肿瘤防治杂志,2021, 28(1): 62-66. DOI:10.16073/j.cnki.cjcpt.2 021.01.09.
Liang YH, Guo AJ, Shi XY, et al. Evaluation of the value of CT in differentiating mucinous adenocarcinoma from tuberculoma in 61 cases of solitary pulmonary solid nodules [J]. Chinese Journal of Cancer Prevention and Treatment, 2021, 28(1):62-66. DOI: 10.16073/j.cnkicjcpt.2021.01.09.
[16]王鹏远, 辛军. 原发性肺黏液腺癌18F-FDG PET/CT影像表现的初步研究[J].实用放射学杂志, 2021, 5(14):749-752. DOI:10.3969/j.issn.1002-1671.2021.05.014.
Wang PY, Xin J. Preliminary study on the 18F-FDG PET/CT imaging manifestations of primary pulmonary mucinous adenocarcinoma [J]. Journal of Practical Radiology, 2021, 5(14): 749-752. DOI: 10.3969/j.issn.1002-1671.2021.05.014.
[17]陈瑶, 孙毅, 石国富, 等.64排螺旋CT对肺孤立性结节的征象分析及诊断价值[J].中华临床医师杂志：电子版,2020,14(10):817-822.DOI:10.3877/cma.j.issn.1674-0785.2020.10.013.
Chen Y, Sun Y, Shi GF, et al. Analysis of signs and diagnostic value of solitary pulmonary nodules by 64 - slice spiral CT [J]. Chinese Journal of Clinical Physicians: Electronic Version, 2020, 14(10): 817-822. DOI: 10.3877/cma.j.issn.1674-0785.2020.10.013.
[18]Shah R, Lim JS, Hum YF, et al. Liquid biopsy in combination with PET/CT versus PET/CT alone in diagnosis of lung cancer in small nodules.[J].J Clin Oncol, 2023, 41(S16):3072-3072. DOI:10.1200/jco.2023.41.16_suppl.3072.
[19]Sung P, Yoon SH, Kim J, et al. Bronchovascular bundle thickening on CT as a predictor of survival and brain metastasis in patients with stage IA peripheral small cell lung cancer[J]. Clin Radiol, 2021, 76(1):37-46. DOI:10.1016/j.crad.2020.08.018.
[20]Lee J, Kim SH, Kang BJ. Prognostic Factors of Disease Recurrence in Breast Cancer Using Quantitative and Qualitative Magnetic Resonance Imaging (MRI) Parameters.[J]. Sci Rep, 2020, 10(1):e7598. DOI: 10.1038/s41598-020-64564-6.
PDF(3525 KB)
京ICP备08002354号-3
