基于机器学习的早期肝肿瘤超声造影诊断模型建立及应用价值

刘丽; 谭鹰; 唐春霖; 郭燕丽

doi:10.16016/j.2097-0927.202307131

您当前的位置：

首页 >

文章列表页 >

基于机器学习的早期肝肿瘤超声造影诊断模型建立及应用价值

临床医学 | 更新时间：2026-01-09

- 基于机器学习的早期肝肿瘤超声造影诊断模型建立及应用价值
- Establishment and application value of a diagnostic model for early hepatic tomors using contrast-enhanced ultrasound based on machine learning
- 陆军军医大学学报 2023年45卷第21期页码：2275-2283
- 作者机构：
  
  1.400038 重庆，陆军军医大学(第三军医大学)：第一附属医院超声科
  2.生物医学工程与影像医学系数字医学教研室
- 作者简介：
  
  [郭燕丽，E-mail：guoyanli71@aliyun.comGUO Yanli, E-mail: guoyanli71@aliyun.com
- 基金信息：
  
  国家国际科技合作专项(2015DFA30920);陆军军医大学第一附属医院重大领域技术创新项目(SWH2016ZDCX4101);国家自然科学基金面上项目(82171956)
- DOI：10.16016/j.2097-0927.202307131
  中图分类号： R319;R730.41;R735.7
- 收稿：2023-07-27，
  
  修回：2023-08-30，
  
  纸质出版：2023-11-15
- 稿件说明：
移动端阅览
刘丽, 谭鹰, 唐春霖, 等. 基于机器学习的早期肝肿瘤超声造影诊断模型建立及应用价值[J]. 陆军军医大学学报, 2023,45(21):2275-2283.

Li LIU, Ying TAN, Chunlin TANG, et al. Establishment and application value of a diagnostic model for early hepatic tomors using contrast-enhanced ultrasound based on machine learning[J]. Journal of Army Medical University, 2023, 45(21): 2275-2283.
刘丽, 谭鹰, 唐春霖, 等. 基于机器学习的早期肝肿瘤超声造影诊断模型建立及应用价值[J]. 陆军军医大学学报, 2023,45(21):2275-2283. DOI： 10.16016/j.2097-0927.202307131.

Li LIU, Ying TAN, Chunlin TANG, et al. Establishment and application value of a diagnostic model for early hepatic tomors using contrast-enhanced ultrasound based on machine learning[J]. Journal of Army Medical University, 2023, 45(21): 2275-2283. DOI： 10.16016/j.2097-0927.202307131.

摘要

目的

建立辅助超声造影(contrast-enhanced ultrasound

CEUS)诊断早期肝肿瘤的机器学习模型

并探索其临床诊断效能。

方法

采用病例-对照研究设计方案

筛选2015年6月至2020年6月在陆军军医大学第一附属医院超声科行CEUS检查

有病理诊断结果且肝肿瘤直径≤30 mm患者的病例资料。最终490例患者共520个肿瘤[恶性肿瘤474个(91.15%)

良性肿瘤46个(8.85%)

]

纳入研究

其中男性406例(82.86%)

女性84例(17.14%); 年龄22~82(51.98±0.46)岁。由4位超声科医师分析常规超声及CEUS影像特征

意见不一致时经讨论达成共识。将纳入病例资料按照近4:1比例随机划分为训练集(

=400)和测试集(

=90)。采用支持向量机、随机森林、邻近算法和Logistic等4种模型分析患者常规超声及CEUS影像特征、慢性肝病史和肿瘤标志物。应用准确性、特异性、敏感性和受试者工作曲线下面积(area under the subject curve

AUC)等指标评价模型性能。

结果

4种模型中

随机森林模型性能最优

在测试集中的准确性、敏感性、特异性和AUC分别为0.97、0.83、0.71、0.987(95%

: 0.934~1.000)

其中AUC与其他3种模型差异无统计学意义(均

0.05)。随机森林模型中按照特征贡献度排名

前5名分别是: 肝硬化、年龄、CEUS和常规超声中直径、延迟期增强方式

与以往研究中得出的临床诊断肝脏恶性肿瘤的指标一致

表明本研究建立的模型具有较好的可解释性。

结论

本研究建立了能联合分析患者CEUS特征、慢性肝病和肿瘤标志物的机器学习模型

在CEUS中诊断早期肝肿瘤的准确性较高。

Abstract

Objective

To establish a machine learning diagnosis model for early liver tumors based on contrast-enhanced ultrasound (CEUS) images and explore its clinical diagnostic efficacy.

Methods

A case-control trial was conducted on the clinical data of the patients with pathologically diagnosed hepatic tumors in a diameter ≤30 mm and undergoing CEUS examinations in the First Affiliated Hospital from June 2015 to June 2020.Finally

a total of 490 patients with 520 liver masses[474 malignant tumors (91.15%) and 46 benign tumors (8.85%)

]

were enrolled

including 406 males (82.86%) and 84 females (17.14%)

with a mean age of 51.98±0.46(22~82) years.Four sonographers were invited to analyze the features of their conventional ultrasonographs and CEUS images

and a consensus needed to be reached through discussion when opinions differ.All enrolled patients were divided into a training set (

=400) and a testing set (

=90) in a ratio of 4:1.Conventional ultrasonographs

CEUS images

history of chronic liver disease and clinical indicators were jointly analyzed using machine learning methods

such as

support vector machines

random forest

nearest neighbor algorithm

and logistic regression.Accuracy

specificity

sensitivity

and area under the receiver operating characteristic curve (AUC) were calculated to evaluate the performance of the models.

Results

Among the 4 models

the random forest performed best

with an accuracy

sensitivity

specificity

and AUC of 0.97

0.83

0.71

and 0.987(95%

: 0.934~1.000)

respectively

in the test set.There was no significant difference in the AUC values in the random forest with the other 3 models (all

0.05).The top 5 contributing features in random forest model were cirrhosis

age

diameters in ultrasound and CEUS

and portal phase enhancement pattern

which were consistent with the clinical diagnostic indicators of malignant liver tumors reported in previous studies

indicating that our established model had a good interp

retability.

Conclusion

Our machine learning model is established based on CEUS features

chronic liver disease and tumor markers

and shows high accuracy in diagnosis of liver tumors in diameter ≤30 mm in CEUS.

关键词

Keywords

references

KOKUDO N, HASEGAWA K, AKAHANE M, et al . Evidence-based clinical practice guidelines for hepatocellular carcinoma: the Japan Society of Hepatology 2013 update (3rd JSH-HCC guidelines)[J ] . Hepatol Res, 2015, 45(2): 123-127. DOI: 10.1111/hepr.12464.

中华人民共和国国家卫生健康委员会医政医管局. 原发性肝癌诊疗指南(2022年版)[J]. 中华消化外科杂志, 2022, 21(2): 143-168. DOI: 10.3760/cma.j.cn115610-20220124-00053.

Bureau of Medical Administration of National Health Commission of the People's Republic of China. Standardization for diagnosis and treatment of hepatocellular carcinoma (2022 edition)[J]. Chin J Dig Surg, 2022, 21(2): 143-168. DOI: 10.3760/cma.j.cn115610-20220124-00053.

WILSON S R, FEINSTEIN S B. Introduction: 4th guidelines and good clinical practice recommendations for contrast enhanced ultrasound (CEUS) in the liver—update 2020 WFUMB in cooperation with EFSUMB, AFSUMB, AIUM and FLAUS[J]. Ultrasound Med Biol, 2020, 46(12): 3483-3484. DOI: 10.1016/j.ultrasmedbio.2020.08.015.

HEIMBACH J K, KULIK L M, FINN R S, et al . AASLD guidelines for the treatment of hepatocellular carcinoma[J ] . Hepatology, 2018, 67(1): 358-380. DOI: 10.1002/hep.29086.

European Association for the Study of the Liver. EASL clinical practice guidelines: management of hepatocellular carcinoma[J]. J Hepatol, 2018, 69(1): 182-236. DOI: 10.1016/j.jhep.2018.03.019.

FAN P L, DING H, MAO F, et al . Enhancement patterns of small hepatocellular carcinoma (≤30 mm) on contrast-enhanced ultrasound: correlation with clinicopathologic characteristics[J ] . Eur J Radiol, 2020, 132: 109341. DOI: 10.1016/j.ejrad.2020.109341.

JANG H J, KIM T K, BURNS P N, et al . Enhancement patterns of hepatocellular carcinoma at contrast-enhanced US: comparison with histologic differentiation[J ] . Radiology, 2007, 244(3): 898-906. DOI: 10.1148/radiol.2443061520.

HUANG J Y, LI J W, LU Q, et al . Diagnostic accuracy of CEUS LI-RADS for the characterization of liver nodules 20 mm or smaller in patients at risk for hepatocellular carcinoma[J ] . Radiology, 2020, 294(2): 329-339. DOI: 10.1148/radiol.2019191086.

LING W W, WANG M N, MA X L, et al . The preliminary application of liver imaging reporting and data system (LI-RADS) with contrast-enhanced ultrasound (CEUS) on small hepatic nodules (≤2 cm)[J ] . J Cancer, 2018, 9(16): 2946-2952. DOI: 10.7150/jca.25539.

刘巧, 曾燕, 王浩林, 等. 机器学习方法对新型冠状病毒肺炎的诊断价值[J]. 中国医学影像学杂志, 2021, 29(4): 293-299. DOI: 10.3969/j.issn.1005-5185.2021.04.002.

LIU Q, ZENG Y, WANG H L, et al . Diagnosis of machine learning methods for COVID-19[J ] . Chin J Med Imaging, 2021, 29(4): 293-299. DOI: 10.3969/j.issn.1005-5185.2021.04.002.

庄锡伟, 黎志全, 刘建雷, 等. 5种机器学习算法在肾结石筛查诊断中的应用及比较[J]. 现代医药卫生, 2023, 39(12): 2045-2048. DOI: 10.3969/j.issn.1009-5519.2023.12.013.

ZHUANG X W, LI Z Q, LIU J L, et al . Application and comparison of five machine learning algorithms in screening and diagnosis of renal calculi[J ] . J Mod Med Health, 2023, 39(12): 2045-2048. DOI: 10.3969/j.issn.1009-5519.2023.12.013.

ALAIMO L, LIMA H A, MOAZZAM Z, et al . Development and validation of a machine-learning model to predict early recurrence of intrahepatic cholangiocarcinoma[J ] . Ann Surg Oncol, 2023, 30(9): 5406-5415. DOI: 10.1245/s10434-023-13636-8.

黄媛, 崔悦, 王冠华, 等. 基于机器学习的肝细胞癌多指标联合诊断模型开发及验证[J]. 中国医学装备, 2023, 20(5): 50-53. DOI: 10.3969/J.ISSN.1672-8270.2023.05.009.

HUANG Y, CUI Y, WANG G H, et al . Development and validation of a model of the combined diagnosis with multiple indexes for HCC on the bases of machine learning[J ] . China Med Equip, 2023, 20(5): 50-53. DOI: 10.3969/J.ISSN.1672-8270.2023.05.009.

韩冰, 遆亚楠, 王蓉, 等. 基于血清标志物的机器学习模型对肝癌诊断的价值评估[J]. 临床检验杂志, 2023, 41(3): 229-234. DOI: 10.13602/j.cnki.jcls.2023.03.16.

HAN B, TI Y N, WANG R, et al . Evaluation of diagnosis value of machine learning model for hepatocellular carcinoma based on serological markers[J ] . Chin J Clin Lab Sci, 2023, 41(3): 229-234. DOI: 10.13602/j.cnki.jcls.2023.03.16.

马凯, 王锦淼, 黄健, 等. 基于平扫CT影像组学模型预测肝血管瘤增强影像的研究[J]. 现代医用影像学, 2023, 32(5): 847-851. DOI: 10.3969/j.issn.1006-7035.2023.05.012.

MA K, WANG J M, HUANG J, et al . Study on prediction of enhanced images of hepatic hemangioma based on the model of plain CT image omics[J ] . Mod Med Imagel, 2023, 32(5): 847-851. DOI: 10.3969/j.issn.1006-7035.2023.05.012.

任闪闪. 应用超声组学术前精准诊断原发性肝癌的多中心研究[D]. 开封: 河南大学, 2022.

REN S S. Multicenter study on accurate diagnosis of primary liver cancer by ultrasound group before academic study[D]. Kaifeng: Henan University, 2022.

CHEN Y D, ZHANG L, ZHOU Z P, et al . Radiomics and nomogram of magnetic resonance imaging for preoperative prediction of microvascular invasion in small hepatocellular carcinoma[J ] . World J Gastroenterol, 2022, 28(31): 4399-4416. DOI: 10.3748/wjg.v28.i31.4399.

SUN K, SHI L T, QIU J F, et al . Multi-phase contrast-enhanced magnetic resonance image-based radiomics-combined machine learning reveals microscopic ultra-early hepatocellular carcinoma lesions[J ] . Eur J Nucl Med Mol Imaging, 2022, 49(8): 2917-2928. DOI: 10.1007/s00259-022-05742-8.

GRIMES D A, SCHULZ K F. An overview of clinical research: the lay of the land[J]. Lancet, 2002, 359(9300): 57-61. DOI: 10.1016/S0140-6736(02)07283-5.

American College of Radiology. ACR CEUS LI-RADS 2016[EB/OL]. [2023-07-01]. https://www.acr.org/Clinical-Resources/Reporting-and-Data-Systems/LI-RADS/CEUS-LI-RADS-v2016.

KIM T K, NOH S Y, WILSON S R, et al . Contrast-enhanced ultrasound (CEUS) liver imaging reporting and data system (LI-RADS) 2017—a review of important differences compared to the CT/MRI system[J ] . Clin Mol Hepatol, 2017, 23(4): 280-289. DOI: 10.3350/cmh.2017.0037.

WILSON S R, LYSHCHIK A, PISCAGLIA F, et al . CEUS LI-RADS: algorithm, implementation, and key differences from CT/MRI[J ] . Abdom Radiol, 2018, 43(1): 127-142. DOI: 10.1007/s00261-017-1250-0.

SHIN J, LEE S, BAE H, et al . Contrast-enhanced ultrasound liver imaging reporting and data system for diagnosing hepatocellular carcinoma: a meta-analysis[J ] . Liver Int, 2020, 40(10): 2345-2352. DOI: 10.1111/liv.14617.

BRAY F, FERLAY J, SOERJOMATARAM I, et al . Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J ] . CA Cancer J Clin, 2018, 68(6): 394-424. DOI: 10.3322/caac.21492.

FORNER A, REIG M, BRUIX J. Hepatocellular carcinoma[J]. Lancet, 2018, 391: 1301-1314. DOI: 10.1016/S0140-6736(18)30010-2.

KUMAR K, KAUR S, KAUR R, et al . The role of contrast-enhanced ultrasound in the evaluation of hepatobiliary lesions and its correlation with pathological findings[ J ] . Pjr, 2022, 87(1): 661-667. DOI: 10.5114/pjr.2022.123563.

KONO Y, LYSHCHIK A, COSGROVE D, et al . Contrast Enhanced Ultrasound (CEUS) Liver Imaging Reporting and Data System (LI-RADS ® ): the official version by the American College of Radiology (ACR)[J ] . Ultraschall Med, 2017, 38(1): 85-86. DOI: 10.1055/s-0042-124369..

ZHENG W, LI Q, ZOU X B, et al .Evaluation of contrast-enhanced US LI-RADS version 2017: application on 2020 liver nodules in patients with hepatitis B infection[J ] . Radiology, 2020, 294(2): 299-307. DOI: 10.1148/radiol.2019190878.

TERZI E, IAVARONE M, POMPILI M, et al . Contrast ultrasound LI-RADS LR-5 identifies hepatocellular carcinoma in cirrhosis in a multicenter restropective study of 1, 006 nodules[J ] . J Hepatol, 2018, 68(3): 485-492. DOI: 10.1016/j.jhep.2017.11.007.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

超声造影肝脏影像报告-数据系统与欧洲超声造影指南对肝脏结节良恶性鉴别的诊断效能对比

基于代谢组学与机器学习筛选高原失血性休克诊断标志物：D-inositol-4-phosphate、Phosalone与Methionine的鉴定与验证

SLC38A1通过促进谷氨酰胺摄取激活PI3K/AKT信号通路驱动肝细胞癌恶性进展

肝细胞癌临床诊治新进展：综合治疗策略的演变