基于多序列MRI的影像组学模型在腮腺肿瘤良恶性诊断中的应用

徐祖良; 盛亮; 何遐遐; 王艳娜; 汪国余

doi:10.16766/j.cnki.issn.1674-4152.002824

基于多序列MRI的影像组学模型在腮腺肿瘤良恶性诊断中的应用

doi: 10.16766/j.cnki.issn.1674-4152.002824

1.
台州市中心医院放射科, 浙江台州 318000
2.
台州市立医院放射科, 浙江台州 318000

基金项目:

浙江省卫生健康科技计划项目 2021KY396

详细信息

通讯作者:
盛亮，E-mail：sl980195@sina.com

中图分类号: R445.2 R739.8
计量
- 文章访问数: 234
- HTML全文浏览量: 29
- PDF下载量: 3
- 被引次数: 0
出版历程
- 收稿日期: 2022-07-12
- 网络出版日期: 2023-04-07

Multi-sequence MRI image data-based machine learning model in the diagnosis of benign and malignant parotid gland tumours

1.
Department of Radiology, Taizhou Central Hospital, Taizhou, Zhejiang 318000, China

摘要

摘要: 目的探究腮腺肿瘤良恶性的诊断中应用基于多序列MRI的影像组学模型的临床价值。方法选取2021年1月1日—2022年5月30日于台州市中心医院就诊经病理检查证实的腮腺肿瘤患者共97例，其中良性肿瘤64例，恶性肿瘤33例，提取患者的临床资料及MRI图像。通过ITK-SNAP软件分割感兴趣区域(ROI)，使用3D-slicer软件的PyRadiomics插件在T₁加权对比增强成像序列(T₁Wce)、T₂加权序列(T₂WI)及基于弥散加权成像(DWI)序列构建的表观弥散系数(ADC)图像中提取120个影像特征，使用Lasso回归进行特征降维，最后使用筛选的影像特征构建支持向量机(SVM)模型。然后绘制ROC曲线，评估模型的诊断效能。结果 2组患者年龄、性别差异无统计学意义。基于患者T₁Wce、T₂WI和ADC序列以及3个序列联合构建4个影像组学模型用于腮腺肿瘤良恶性诊断，AUC分别为T₁Wce模型0.752，T₂WI模型0.776，ADC模型0.810，T₁Wce+T₂WI+ADC联合模型0.897，三序列联合模型的AUC显著高于单个序列模型。结论基于MRI影像数据构建的影像组学模型能够有效用于辅助腮腺肿瘤的良恶性诊断，其中联合T₁Wce、T₂WI和ADC序列构建的模型具有最佳的诊断效能。
- 腮腺肿瘤 /
- 磁共振成像 /
- 影像组学 /
- 机器学习 /
- 支持向量机
Abstract: Objective The clinical value of multi-sequence MRI image data-based radiomics in the diagnosed of benign and malignant parotid gland tumors. Methods A total of 97 patients with parotid gland tumours diagnosed by pathological examination in Taizhou Central Hospital from January 1, 2021 to May 30, 2022 were selected, including 64 benign tumours and 33 malignant tumours. Meantime, patients' clinical data and MRI images were extracted. ITK-SNAP was used to segment regions of interest. In addition, PyRadiomics plugin of 3D-Slicer was used to extract 120 image features from T₁-weighted contrast enhancement images (T₁Wce), T₂-weighted images (T₂WI) and apparent diffusion coefficient (ADC) images that were constructed on the basis of diffusion-weighted imaging (DWI) sequence. Then, Lasso regression was used for the reduction of image features. Finally, a support vector machine model (SVM) was constructed using the selected image features. Then, the ROC curve was drawn to evaluate the diagnostic efficiency of each model. Results No significant differences in age and gender ratio was observed between the two groups. Four imaging models were constructed, including the models based on T₁Wce, T₂WI, ADC images and the combination of the three sequences. The AUC was as follows: T₁Wce model 0.752, T₂WI model 0.776, ADC model 0.810, T₁Wce+T₂WI+ADC model 0.897. The AUC of the three-sequence combination model was significantly higher than that of the single-sequence model. Conclusion The radiomics model based on MRI image data can be effectively used for assisting the diagnosis of benign and malignant parotid gland tumours. Moreover, the model based on the combination of T₁Wce, T₂WI and ADC images has the best diagnostic efficiency.
- Parotid gland tumour /
- Magnetic resonance imaging /
- Radiomics /
- Machine learning /
- Support vector machine

HTML全文

图 1 良恶性腮腺肿瘤MRI各序列图像

Figure 1. The multi-sequence MRI images of benign and malignant parotid gland tumors

下载: 全尺寸图片幻灯片

图 2 各序列SVM模型的ROC曲线

Figure 2. The ROC curve of SVM model for each sequence

下载: 全尺寸图片幻灯片

表 1 各序列SVM模型ROC曲线分析结果

Table 1. The results of ROC curve analysis for SVM model of each sequence

模型	AUC	95% CI	灵敏度(%)	特异度(%)
T₁Wce	0.752^a	0.654~0.834	48.48	1.00
T₂WI	0.776^a	0.680~0.854	75.76	71.81
ADC	0.810^a	0.717~0.882	60.61	87.50
T₁Wce+T₂WI+ADC	0.897	0.819~0.950	81.82	93.75
注：与T₁Wce+T₂WI+ADC模型比较，^aP < 0.05。

下载: 导出CSV

表 2 T₁Wce+T₂WI+ADC联合模型的影像学特征

Table 2. The imaging features of T₁Wce+T₂WI+ADC combined model

序列	特征	特征类别
T₁Wce	球形不对称	3D形状特征
	差分方差	GLCM特征
	高灰度级运行强化	GLRLM特征
T₂WI	低依赖强化	GLDM特征
	熵依赖	GLDM特征
	混杂	NGTDM特征
ADC	广域强化	GLSZM特征
	区域大小不均匀	GLSZM特征
	表面体积比	3D形状特征
	高灰度级区域强化	GLSZM特征

下载: 导出CSV

参考文献(19)

[1]	江晓勇, 杨志辉, 陈希希, 等. 不同类型腮腺肿瘤的临床特点及远期复发情况分析[J]. 中华全科医学, 2019, 17(5): 790-792, 796. doi: 10.16766/j.cnki.issn.1674-4152.000793 JIANG X Y, YANG Z H, CHEN X X, et al. Clinical characteristics and long-term recurrence of different types of parotid gland tumors[J]. Chinese Journal of General Practice, 2019, 17(5): 790-792, 796. doi: 10.16766/j.cnki.issn.1674-4152.000793
[2]	唐晨虎, 马东, 孙伟, 等. CT和MRI在腮腺良、恶性肿瘤鉴别诊断中的价值[J]. 中国临床医学影像杂志, 2021, 32(2): 90-93, 103. https://www.cnki.com.cn/Article/CJFDTOTAL-LYYX202102007.htm TANG C H, MA D, SUN W, et al. Value of CT and MRI in differential diagnosis of benign and malignant parotid gland tumors[J]. Journal of China Clinic Medical Imaging, 2021, 32(2): 90-93, 103. https://www.cnki.com.cn/Article/CJFDTOTAL-LYYX202102007.htm
[3]	张世坤, 王智明. 腮腺肿瘤诊断方法研究进展[J]. 中华实用诊断与治疗杂志, 2018, 32(4): 411-413. https://www.cnki.com.cn/Article/CJFDTOTAL-HNZD201804031.htm ZHANG S K, WANG Z M. Progress in the diagnosis of parotid gland tumors[J]. Chinese Journal of Practical Diagnosis and Therapy, 2018, 32(4): 411-413. https://www.cnki.com.cn/Article/CJFDTOTAL-HNZD201804031.htm
[4]	韩蕾, 邬小平. 磁共振成像技术在腮腺肿瘤诊断中的研究进展[J]. 磁共振成像, 2022, 13(2): 133-136. https://www.cnki.com.cn/Article/CJFDTOTAL-CGZC202202032.htm HAN L, WU X P. Research progress of magnetic resonance imaging in the diagnosis of parotid gland tumors[J]. Magnetic Resonance Imaging, 2022, 13(2): 133-136. https://www.cnki.com.cn/Article/CJFDTOTAL-CGZC202202032.htm
[5]	朱凌, 王灿, 杨功鑫, 等. 腮腺肿瘤性病变的影像学诊断思路及功能MRI价值评估[J]. 中华放射学杂志, 2022, 56(5): 587-590. ZHU L, WANG C, YANG G X, et al. Imaging diagnosis of parotid neoplastic lesions and evaluation of functional MRI value. [J]. Chinese Journal of Radiology, 2022, 56(5): 587-590.
[6]	LITJENS G, KOOI T, BEJNORDI B E, et al. A survey on deep learning in medical image analysis[J]. Med Image Anal, 2017, 42: 60-88.
[7]	BINCZYK F, PRAZUCH W, BOZEK P, et al. Radiomics and artificial intelligence in lung cancer screening[J]. Transl Lung Cancer Res, 2021, 10(2): 1186-1199.
[8]	LIBERINI V, LAUDICELLA R, BALMA M, et al. Radiomics and artificial intelligence in prostate cancer: new tools for molecular hybrid imaging and theragnostics[J]. Eur Radiol Exp, 2022, 6(1): 27.
[9]	LI Q, JIANG T, ZHANG C, et al. A nomogram based on clinical information, conventional ultrasound and radiomics improves prediction of malignant parotid gland lesions[J]. Cancer Lett, 2022, 527: 107-114.
[10]	ZHENG Y M, XU W J, HAO D P, et al. A CT-based radiomics nomogram for differentiation of lympho-associated benign and malignant lesions of the parotid gland[J]. Eur Radiol, 2021, 31(5): 2886-2895.
[11]	PILUDU F, MARZI S, RAVANELLI M, et al. MRI-based radiomics to differentiate between benign and malignant parotid tumors with external validation[J]. Front Oncol, 2021, 11: 656918. DOI: 10.3389/fonc.2021.656918.
[12]	RAJKOMAR A, DEAN J, KOHANE I. Machine learning in medicine[J]. N Engl J Med, 2019, 380(14): 1347-1358.
[13]	JUAN C J, HUANG T Y, LIU Y J, et al. Improving diagnosing performance for malignant parotid gland tumors using machine learning with multifeatures based on diffusion-weighted magnetic resonance imaging[J]. NMR Biomed, 2022, 35(3): e4642. DOI: 10.1002/nbm.4642.
[14]	YUSHKEVICH P A, PIVEN J, HAZLETT H C, et al. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability[J]. Neuroimage, 2006, 31(3): 1116-1128.
[15]	VAN GRIETHUYSEN J, FEDOROV A, PARMAR C, et al. Computational radiomics system to decode the radiographic phenotype[J]. Cancer Res, 2017, 77(21): e104-e107.
[16]	FRIEDMAN J, HASTIE T, TIBSHIRANI R. Regularization paths for generalized linear models via coordinate descent[J]. J Stat Softw, 2010, 33(1): 1-22.
[17]	程顺达, 程颖, 孙士江. 基于机器学习的肿瘤智能辅助诊断方法[J]. 电子科技, 2022, 35(5): 56-59. https://www.cnki.com.cn/Article/CJFDTOTAL-DZKK202205009.htm CHENG S D, CHENG Y, SUN S J. Intelligent auxiliary diagnosis of tumor based on machine learning[J]. Electronics Technology, 2022, 35(5): 56-59. https://www.cnki.com.cn/Article/CJFDTOTAL-DZKK202205009.htm
[18]	杜文, 彭歆. 人工智能技术在口腔颌面部肿瘤诊断中应用的研究进展[J]. 口腔疾病防治, 2022, 30(5): 361-365. https://www.cnki.com.cn/Article/CJFDTOTAL-GDYB202205009.htm DU W, PENG Q. Research progress of artificial intelligence technology in the diagnosis of oral and maxillofacial tumors[J]. Prevention and Treatment of Oral Diseases, 2022, 30(5): 361-365. https://www.cnki.com.cn/Article/CJFDTOTAL-GDYB202205009.htm
[19]	冯利, 岳小飞. 9种结合式机器学习算法在肿瘤早期诊断中的准确性比较研究[J]. 医学信息学杂志, 2022, 43(5): 25-29. https://www.cnki.com.cn/Article/CJFDTOTAL-YXQB202205008.htm FENG L, YUE X F. Comparative study on the accuracy of 9 combined machine learning algorithms in early diagnosis of tumor[J]. Journal of Medical Informatics, 2022, 43(5): 25-29. https://www.cnki.com.cn/Article/CJFDTOTAL-YXQB202205008.htm