[1]王毅,李远哲,李淑婷,等.基于多模态磁共振放射组学与临床指标的前列腺癌智能检测及风险预测模型建立[J].中国医学物理学杂志,2023,40(2):251-260.[doi:DOI:10.3969/j.issn.1005-202X.2023.02.021]
　WANG Yi,LI Yuanzhe,LI Shuting,et al.Establishment of models for the intelligent detection and risk prediction of prostate cancer based on the combination of multi-modality magnetic resonance imaging radiomics and clinical indicators[J].Chinese Journal of Medical Physics,2023,40(2):251-260.[doi:DOI:10.3969/j.issn.1005-202X.2023.02.021]
点击复制

基于多模态磁共振放射组学与临床指标的前列腺癌智能检测及风险预测模型建立()
分享到：

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
40卷
期数:
2023年第2期
页码:
251-260
栏目:
医学人工智能
出版日期:
2023-03-03

文章信息/Info

Title:
Establishment of models for the intelligent detection and risk prediction of prostate cancer based on the combination of multi-modality magnetic resonance imaging radiomics and clinical indicators
文章编号:
1005-202X（2023）02-0251-10
作者:
王毅; 李远哲; 李淑婷; 赖清泉
福建医科大学附属第二医院CT/MRI室， 福建 泉州 362000
Author(s):
WANG Yi;  LI Yuanzhe;  LI Shuting;  LAI Qingquan
Department of CT/MRI, the Second Affiliated Hospital, Fujian Medical University, Quanzhou 362000, China
关键词:
前列腺癌; 放射组学; 磁共振; 多模态机器学习
Keywords:
Keywords: prostate cancer radiomics magnetic resonance imaging multi-modality machine learning
分类号:
R318；R737.5
DOI:
DOI:10.3969/j.issn.1005-202X.2023.02.021
文献标志码:
A
摘要:
目的：利用多模态磁共振放射组学开发前列腺癌自动检测模型，并使用列线图构建多因素回归模型，将前列腺MRI放射组学特征与临床多个检测指标进行整合，从而对患前列腺癌风险性进行预测。方法：回顾性研究于2019年2月~2021年10月病理证实为前列腺癌和其他前列腺良性肿瘤的患者133例。所有病例均行前列腺直肠指检（DRE）、前列腺特异性抗原（PSA）、游离前列腺特异性抗原（F-PSA）、FPSA/PSA检测。治疗前多模态前列腺MRI图像（DWI+DCE+T2WI）用于提取放射特征，最大相关最小冗余（mRMR）算法用于消除混杂变量，使用最小绝对收缩和选择算子（LASSO）逻辑回归进行放射特征选择。通过曲线下面积（AUC）、准确性、特异性、敏感性评估放射特征的诊断性能；通过多元logistic回归选择临床指标和放射组学特征模型来制定放射组学列线图，并使用校准曲线和Hosmer-lemeshow试验验证其可靠性。结果：两名观察者测量的所有数据ICC均在0.80以上。所有前列腺MRI图像随机分为训练组和验证组（7：3）。在训练组中，DWI、DCE和T2WI的AUC分别为0.882、0.821、0.848，在验证组中，DWI、DCE和T2WI的AUC分别为0.861、0.810、0.838；三模态联合模型的训练组和验证组AUC分别为0.912、0.898。Delong检验结果显示DWI模型性能优于DCE和T2WI模型性能，DCE和T2WI模型检测性能相仿，三模态联合模型性能优于任意一种模型性能。使用ROC曲线评估列线图、影像组学和临床指标的预测性能，结果显示列线图的AUC值为0.941，准确率、敏感性、特异性分别为0.929、0.891、0.893。列线图前列腺癌预测性能最好，临床指标的预测性能较差，校准曲线和Hosmer-Lemeshow检验结果也验证了上述观点。结论：多模态前列腺MRI放射组学模型能准确鉴别前列腺肿瘤的良恶性，放射组学列线图在前列腺癌风险预测中表现出令人满意的效果。
Abstract:
Abstract: Objective To develop an automatic detection model of prostate cancer using multi-modality magnetic resonance imaging (MRI) radiomics, and to predict the risk of prostate cancer by a multifactor regression model constructed by nomogram based on the combination of prostate MRI radiomics and clinical indicators. Methods A retrospective analysis was conducted on 133 patients with prostate cancer and other benign prostatic lesions confirmed by pathology from February 2019 to October 2021. All patients underwent directeral rectum examination (DRE), and were tested for prostate specific antigen (PSA), free-prostate specific antigen (F-PSA) and F-PSA/PSA. After extracting radiological features from multi-modality prostate MRI images (DWI+DCE+T2WI) before treatment, the minimal redundancy maximal relevance (mRMR) algorithm was used for eliminating hybrid variables, and the least absolute shrinkage and selection operator (LASSO) for radiological feature selection. The diagnostic performances of radiological features were evaluated by area under ROC curve (AUC), accuracy, specificity and sensitivity. Multiple logistic regression analysis was used to select clinical indicators which were then combined with radiomics feature model to formulate radiomics nomogram. The model reliability was verified by calibration curve and Hosmer-lemeshow test. Results The ICC of all data measured by two observers was above 0.80. All MRI images of the prostate were randomly divided into training group and verification group at a ratio of 7:3. The AUC of DWI, DCE and T2WI were 0.882, 0.821, 0.848 in training group, and 0.861, 0.810, 0.838 in verification group, while the combination model of triple-modality MRI achieved AUC of 0.912 and 0.898 in training group and validation group, respectively. The Delong test results show that DWI model outperformed DCE and T2WI models (the latter two had similar performances), and that the performance of combination model of triple-modality MRI was superior to that of any other model. ROC curve was used to evaluate the predictive performance of nomogram, radicomics and clinical indicators, and the results revealed that the AUC, accuracy, sensitivity and specificity of nomogram were 0.941, 0.929, 0.891 and 0.893, respectively. Nomogram had the best predictive performance for prostate cancer, and the predictive performance of clinical factors was poor. Both calibration curve and Hosmer-lemeshow test results verified the above findings. Conclusion Multi-modality prostate MRI radiomics model can accurately identify benign and malignant prostate tumors. Radiomics nomogram shows a satisfactory performance in the prediction of prostate cancer risk.

相似文献/References:

[1]董晓庆,胡杰,陆春花,等.前列腺癌图像引导放射治疗精准度评估[J].中国医学物理学杂志,2016,33(7):658.[doi:10.3969/j.issn.1005-202X.2016.07.003]
　[J].Chinese Journal of Medical Physics,2016,33(2):658.[doi:10.3969/j.issn.1005-202X.2016.07.003]
[2]倪千喜,Pei-Fong WONG,张九堂.不同治疗床在调强放射治疗和容积旋转放射治疗中的剂量影响分析[J].中国医学物理学杂志,2016,33(11):1116.[doi:10.3969/j.issn.1005-202X.2016.11.007]
　[J].Chinese Journal of Medical Physics,2016,33(2):1116.[doi:10.3969/j.issn.1005-202X.2016.11.007]
[3]曹洋森,于春山,孙永健,等.Monaco两种优化模式在前列腺癌容积旋转调强中的剂量学比较[J].中国医学物理学杂志,2016,33(11):1126.[doi:10.3969/j.issn.1005-202X.2016.11.009]
　[J].Chinese Journal of Medical Physics,2016,33(2):1126.[doi:10.3969/j.issn.1005-202X.2016.11.009]
[4]鞠孟阳,史玉静,李金凯,等. 直肠内气腔对前列腺癌放疗计划的影响[J].中国医学物理学杂志,2018,35(8):879.[doi:DOI:10.3969/j.issn.1005-202X.2018.08.003]
　JU Mengyang,SHI Yujing,LI Jinkai,et al. Effects of gas cavity in rectum on radiotherapy plan for prostate cancer[J].Chinese Journal of Medical Physics,2018,35(2):879.[doi:DOI:10.3969/j.issn.1005-202X.2018.08.003]
[5]张君席,刘振湘,张平,等. 基于太赫兹光谱和成像的前列腺癌检测[J].中国医学物理学杂志,2019,36(5):556.[doi:DOI:10.3969/j.issn.1005-202X.2019.05.012]
　ZHANG Junxi,LIU Zhenxiang,ZHANG Ping,et al. Detection of prostate cancer based on terahertz spectroscopy and imaging[J].Chinese Journal of Medical Physics,2019,36(2):556.[doi:DOI:10.3969/j.issn.1005-202X.2019.05.012]
[6]张祥斌,李光俊,张英杰,等. 基于形变配准算法评估前列腺癌分次间剂量的可行性研究[J].中国医学物理学杂志,2019,36(9):995.[doi:DOI:10.3969/j.issn.1005-202X.2019.09.001]
　ZHANG Xiangbin,LI Guangjun,ZHANG Yingjie,et al. Feasibility of deformable image registration algorithm for interfractional dose calculation in radiotherapy for prostate cancer[J].Chinese Journal of Medical Physics,2019,36(2):995.[doi:DOI:10.3969/j.issn.1005-202X.2019.09.001]
[7]徐伟,杨涛,解传滨,等. 质子点扫描技术与光子Tomotherapy在前列腺癌应用中的剂量学比较[J].中国医学物理学杂志,2019,36(10):1129.[doi:DOI:10.3969/j.issn.1005-202X.2019.10.003]
　XU Wei,YANG Tao,XIE Chuanbin,et al. Dosimetric comparison between single spot scanning of proton therapy and tomotherapy for prostate cancer[J].Chinese Journal of Medical Physics,2019,36(2):1129.[doi:DOI:10.3969/j.issn.1005-202X.2019.10.003]
[8]方春锋,解传滨,徐寿平,等.不同铅门模式下高危分期前列腺癌螺旋断层放射治疗剂量学比较[J].中国医学物理学杂志,2020,37(2):138.[doi:DOI:10.3969/j.issn.1005-202X.2020.02.002]
　FANG Chunfeng,XIE Chuanbin,et al.Dosimetric study of helical tomotherapy for high-risk prostate cancer under different jaw modes[J].Chinese Journal of Medical Physics,2020,37(2):138.[doi:DOI:10.3969/j.issn.1005-202X.2020.02.002]
[9]田龙,闫洁诚,李明辉,等.利用多次采集计划CT和锥形束CT评价前列腺癌靶区运动相关性[J].中国医学物理学杂志,2021,38(2):172.[doi:DOI:10.3969/j.issn.1005-202X.2021.02.009]
　TIAN Long,YAN Jiecheng,LI Minghui,et al.Evaluating the correlation between pre- and in-treatment target displacements in prostate cancer radiotherapy by multiple acquisition planning CT and cone-beam CT[J].Chinese Journal of Medical Physics,2021,38(2):172.[doi:DOI:10.3969/j.issn.1005-202X.2021.02.009]
[10]潘兴晨,汪冬,胡丽琴.基于靶区与危及器官重叠度的自动治疗计划设计[J].中国医学物理学杂志,2021,38(10):1203.[doi:DOI:10.3969/j.issn.1005-202X.2021.10.004]
　PAN Xingchen,WANG Dong,HU Liqin.Automatic treatment planning based on the degree of overlap between OAR and PTV[J].Chinese Journal of Medical Physics,2021,38(2):1203.[doi:DOI:10.3969/j.issn.1005-202X.2021.10.004]

备注/Memo

备注/Memo:
【收稿日期】2022-11-05 【基金项目】福建省卫生健康科技计划（2020QNA059, 2021QNA038） 【作者简介】王毅，硕士，主治医师，研究方向：骨骼肌肉、肿瘤MRI成像技术与诊断、人工智能，E-mail: fjmuwang@sina.cn 【通信作者】赖清泉，主任医师，副教授，硕士生导师，研究方向：骨骼肌肉、肿瘤MRI成像技术与诊断、人工智能，E-mail: laiqingquan888@163.com
更新日期/Last Update: 2023-03-03