肺腺癌亚型分类技术研究进展

doi:10.3778/j.issn.1002-8331.2210-0116

摘要/Abstract

摘要： 肺腺癌存在多种不同类型，各有表征，准确对其分类是临床诊断和治疗的重要依据。从肺腺癌组织病理学、影像学、基因组学等多个方面进行肺腺癌亚型分类研究一直是临床研究的热点问题之一。特别是近年来机器学习和深度学习技术的发展为肺腺癌分类研究提供了新的方法和思路。详细阐述了当前肺腺癌分类技术的研究进展，对各种亚型分类技术应用进展进行了系统的评价。总结了各类分类技术的优缺点、传统分类方法的难易程度和常用的机器学习、深度学习技术模型的算法复杂度，分析了当前研究的相关问题，并对未来的研究方向进行了展望。

关键词: 肺腺癌, 分类技术, 计算机断层扫描（CT）影像, 病理组织图像, 机器学习, 深度学习

Abstract: There are many different types of lung adenocarcinoma, and each has its own characteristics. Accurate classification of lung adenocarcinoma is an important basis for clinical diagnosis and treatment. Classification of lung adenocarcinoma subtypes from histopathology, imaging, genomics and other aspects has always been one of the hot issues in clinical research. Especially in recent years, the development of machine learning and deep learning technology has provided new methods and ideas for the classification of lung adenocarcinoma. In this paper, the research progress of the classification techniques of lung adenocarcinoma is described in detail, and the application progress of the classification techniques of various subtypes is systematically evaluated. This paper summarizes the advantages and disadvantages of various classification technologies, the difficulty of traditional classification methods and the algorithm complexity of common machine learning and deep learning technology models, analyzes the relevant problems in current research, and looks forward to the future research direction.

Key words: lung adenocarcinoma, classification technique, computed tomography（CT） imaging, pathological tissue image, machine learning, deep learning

刘茗传, 张魁星, 江梅, 张晓丽, 李丽萍. 肺腺癌亚型分类技术研究进展[J]. 计算机工程与应用, 2023, 59(17): 67-79.

LIU Mingchuan, ZHANG Kuixing, JIANG Mei, ZHANG Xiaoli, LI Liping. Advances in Classification of Lung Adenocarcinoma Subtypes[J]. Computer Engineering and Applications, 2023, 59(17): 67-79.

参考文献

[1] CRONIN K A，LAKE，A J，SCOTT，S，et al.Annual report to the nation on the status of cancer，part I：national cancer statistics[J].Cancera Journal of the American Cancer Society，2018，124（13）：2785-2800.
[2] MARX A，CHAN J K，COINDRE J M，et al.The 2015 WHO classification of tumors of the thymus：continuity and changes[J].Journal of Thoracic Oncology，2015，10（10）：1383.
[3] HUO J W，HUANG X T，LI X，et al.Pneu-moictype lung adenocarcinoma with different ranges exhibiting different clinical，imaging，and pathological characteristics[J].Insights Into Imaging，2021，12（1）：169.
[4] 徐庆东，蔡庆.不同密度早期肺腺癌影像特征与病理学分类的关系研究[J].影像研究与医学应用，2021，5（23）：57-58.
XU Q D，CAI Q.Study on the relationship between imaging features and pathological classification of early lung adenocarcinoma with different densities[J].Journal of Imaging Research and Medical Applications，2021，5（23）：57-58.
[5] 戴书华，刘国芳，向东生.肺磨玻璃结节CT值测量在早期癌症诊断中的意义[J].中华肺部疾病杂志，2019，12（6）：770-771.
DAI S H，LIU G F，XIANG D S.Significance of CT value measurement of pulmonary ground glass nodules in early cancer diagnosis[J].Chinese Journal of Lung Diseases，2019，12（6）：770-771.
[6] 龚晨虎，徐辉.肺磨玻璃结节行CT检查对诊断早期肺腺癌的临床价值研究[J].当代医学，2021，27（31）：172-173.
GONG C H，XU H.Clinical value of CT examination for lung ground glass nodules in the diagnosis of early lung adenocarcinoma[J].Contemporary Medicine，2021，27（31）：172-173.
[7] 张睿娟，雷弋.早期肺腺癌临床病理特征及分子特征研究进展[J].现代肿瘤医学，2021，29（23）：4246-4250.
ZHANG R J，LEI Y.Research progress on clinicopathological and molecular characteristics of early lung adenocarcinoma[J].Journal of Modern Oncology，2021，9（23）：4246-4250.
[8] 程春，刘郁鹏，谢鹏飞，等.早期肺腺癌薄层CT影像学特征诊断病理学类型的价值[J].交通医学，2022，36（1）：15-18.
CHENG C，LIU Y P，XIE P F，et al.The value of thin slice CT imaging features in the diagnosis of pathological types in early lung adenocarcinoma[J].Medical Journal of Communications，2022，36（1）：15-18.
[9] CLARK K，VENDT B，SMITH K，et al.The cancer imaging archive（TCIA）：maintaining and operating a public information repository[J].Journal of Digital Imaging，2013，26（6）：1045-1057.
[10] ZHENG J，YANG D，ZHU Y，et al.Pulmonary nodule risk classification in adenocarcinoma from CT images using deep CNN with scale transfer module[J].IET Image Processing，2020，14（8）：1481-1489.
[11] BORKOWSKI A A，BUI M M，THOMAS L B，et al.Lung and colon cancer histopathological image dataset（LC25000）[J].arXiv：1912.12142v1，2019.
[12] 周志华.机器学习[M].北京：清华大学出版社，2016：29-35.
ZHOU Z H.Machine learning[M].Beijing：Tsinghua University Press，2016：29-35.
[13] DIAYRUL D.Chest CT scan image lung[DB/OL].Kaggle Datasets，2022[2022-10-10].https：//www.kaggle.com/diayruldip.
[14] HU B，REN W，FENG Z，et al.Correlation between CT imaging characteristics and pathological diagnosis for subcentimeter pulmonary nodules[J].Thorac Cancer，2022，13（7）：1067-1075.
[15] 刘沁峰，刘伟军，王云鹏，等.基于医学工程学的放射组学研究应用探讨[J].中国医疗设备，2019，34（11）：165-168.
LIU Q F，LIU W J，WANG Y P，et al.Research and application of radiomics based onmedical engineering[J].China Medical Devices，2019，34（11）：165-168.
[16] 王梅，曹捍波，许华权.MSCT对最大径≤1?cm肺腺癌亚型分型的诊断价值[J].医学影像学杂志，2017（8）：1466-1470.
WANG M，CAO H B，XU H Q.The diagnostic value of MSCT in subtyping of lung adenocarcinoma with maximum diameter ≤1?cm[J].Journal of Medical Imaging，2017（8）：1466-1470.
[17] SI M J，TAO X F，DU G Y，et al.Thin-section computed tomography-histopathologic comparisons of pulmonary focal interstitial fibrosis，atypical adenomatous hyperplasia，adenocarcinoma in situ，and minimally invasive aden-ocarcinoma with pure ground-glass opacity[J].European Journal of Radiology，2016，85（10）：1708-1715.
[18] LIU Y，SUN H，ZHOU F，et al.Imaging features of TSCT predict the classification of pulmonary preinvasive lesion，minimally and invasive adenocarcinoma presented as ground glass nodules[J].Lung Cancer，2017，108：192-197.
[19] YANAGAWA M，JOHKOH T，NOGUCHI M，et al.Radiological prediction of tumor invasiveness of lung adenocarcinoma on thin-section CT[J].Medicine，2017，96（11）：e6331.
[20] KOYAMA H，OHNO Y，AOYAMA N，et al.Comparison of STIR turbo SE imaging and diffusion-weighted imaging of the lung：capability for detection and subtype classification of pulmonary adenocarcinomas[J].International Journal of Medical Radiology，2010，20（4）：790-800.
[21] GONG J，LIU J Y，HAO W，et al.Computer-aided diagnosis of ground-glass opacity pulmonary nodules using radiomic features analysis[J].Physics in Medicine and Biology，2019，64（13）：135015.
[22] LI X H，ZHANG W，DU D D，et al.CT features and quantitative analysis of subsolid nodule lung adenocarcinoma for pathological classification prediction[J].Chinese Journal of Medical Imaging，2018，26（9）：658-663.
[23] YANG S M，CHEN L W，WANG H J，et al.Extraction of radiomic values from lung adenocarcinoma with near-pure subtypes in the international association for the study of Lung Cancer/the American Thoracic Society/the European Respiratory Society（IA-SLC/ATS/E-RS） classification[J].Lung Cancer Journal of the International Association for the Study of Lung Cancer，2018，119：56-63.
[24] YANG B，GUO L L，LU G M，et al.Radiomic signature：a non-invasive biomarker for discriminating invasive and non-invasive cases of lung adenocarcinoma[J].Cancer Management and Research，2019，11：7825-7834.
[25] LIU J，CUI J J，LIU F，et al.Multi-subtype classification model for non-small cell lung cancer based on radiomics：SLS model[J].Medical Physics，2019，46（7）：3091-3100.
[26] HUANG Z W，LYU M，AI Z，et al.Pre-operative prediction of Ki-67 expression in various histological subtypes of lung adenocarcinoma based on CT radiomic features[J].Frontiers in Surgery，2021，8：457-470.
[27] HAN Y，MA Y，WU Z，et al.Histologic subtype classification of non-small cell lung cancer using PET/CT images[J].European Journal of Nuclear Medicine and Molecular Imaging，2021，48：350-360.
[28] WANG S，WANG R，ZHANG S，et al.3D convolutional neural network for differentiating pre-invasive lesions from invasive adenocarcinomas appearing as ground-glass nodules with diameters ≤3?cm using HRCT[J].Quanti-tative Imaging in Medicine & Surgery，2018，8（5）：491-499.
[29] NI Y F，YANG Y Y，ZHENG D Z，et al.The invasiveness classification of ground-glass nodules using 3D attention network and HRCT[J].Journal of Digital Imaging，2020，33（5）：1144-1154.
[30] 骆源，徐启飞，吕泽政，等.三维ResNet网络预测肺腺癌结节亚型的效能及其稳定性[J].天津医科大学学报，2022，28（3）：295-300.
LUO Y，XU Q F，LV Z Z，et al.Efficacy and stability of 3D ResNet network in predicting pulmonary adenocarcinoma nodule subtypes[J].Journal of Tianjin Medical University，2022，28（3）：295-300.
[31] LV Y，WEI Y，XU K，et al.3D deep learning versus the current methods for predicting tumor invasiveness of lung adenocarcinoma based on high-resolution computed tomography images[J].Frontiers in Oncology，2022，12：995870.
[32] GONG J，LIU J，LI H，et al.Deep learning based stage-wise risk stratification for early lung adenocarcinoma in CT images：a multicenter study[J].Cancers（Basel），2021，13（13）：3300.
[33] WANG X，LI Q，CAI J，et al.Predicting the invasiveness of lung adenocarcinomas appearing as ground-glass nodule on CT scan using multi-task learning and deep radiomics[J].AME Publishing Company，2020，9（4）：1397-1406.
[34] WANG C D，SHAO J，LV J W，et al.Deep learning for predicting subtype classification and survival of lung adenocarcinoma on computed tomography[J].Translational Oncology，2021，14（8）：101141.
[35] CHO H，LEE H Y，KIM E，et al.Radiomics guided deep neural networks stratify lung adenocarcinoma prognosis from CT scans[J].Communications Biology，2021，4（1）：1286.
[36] DING H，XIA W，ZHANG L，et al.CT-based deep learning model for invasiveness classification and micropapillary pattern prediction within lung adenocarcinoma[J].Frontiers in Oncology，2020，10：1186-1195.
[37] 杨婧，耿辰，王海林，等.基于DenseNet的低分辨CT影像肺腺癌组织学亚型分类[J].浙江大学学报（工学版），2019，53（6）：1164-1170.
YANG J，GENG C，WANG H L，et al.Classification of histological subtypes of lung adenocarcinoma on low-resolution CT images based on DenseNet[J].Journal of Zhejiang University（Engineering Science），2019，53（6）：1164-1170.
[38] ALI I，MUZAMMIL M，HAQ I U，et al.Deep feature selection and decision level fusion for lungs nodule classification[J].IEEE Access，2021，9：18962-18973.
[39] FU Y，XUE P，LI N，et al.Fusion of 3D lung CT and serum biomarkers for diagnosis of multiple pathological types on pulmonary nodules[J].Computer Methods and Programs in Biomedicine，2021，210：106381-106393.
[40] LIU H，JIAO Z C，HAN W J，et al.Identifying the histologic subtypes of non-small cell lung cancer with computed tomography imaging：a comparative study of capsule net，convolutional neural network，and radiomics[J].Quantitative Imaging in Medicine and Surgery，2021，11（6）：2756-2765.
[41] 医学名词审定委员会结核病学名词审定分委员会.结核病学名词[M].北京：科学出版社，2019.
Subcommittee on the Validation of Medical Terms in Tuberculosis.Terminology in tuberculosis[M].Beijing：Science Press，2019.
[42] TRAVIS W D，BRAMBILLA E，NICHOLSON A G，et al.The 2015 World Health organization classification of lung tumors impact of genetic，clinical and radiologic advances since the 2004 classification[J].Journal of Thor-acic Oncology，2015，10（9）：1243-1260.
[43] PENDAR A，BEHZAD H，ALIREZA M E，et al.Representation learning-based unsupervised domain adaptation for classification of breast cancer histopathology images[J].Biocybernetics and Biomedical Engineering，2018，38（3）：671-683．
[44] 王奎，张宏毅，庞瑶，等.基于快速诊断工具建立的肺腺癌亚型研究进展[J/OL].中国胸心血管外科临床杂志：1-7[2022-08-10].http：//kns.cnki.net/kcms/detail/51.1492.R.
20220329.1755.008.html.
WANG K，ZHANG H Y，PANG Y，et al.Research progress of lung adenocarcinoma subtypes based on rapid diagnostic tools[J/OL].Chinese Journal of Clinical Thoracic and Cardiovascular Surgery：1-7[2022-08-10].http：//kns.cnki.net/kcms/detail/51.1492.R.20220329.1755.008.html.
[45] TRAVIS W D.Pathology of lung cancer[J].Clinics in Chest Medicine，2011，32（4）：669-692.
[46] SCHMID K，ANGERSTEIN N，GELEFF S，et al.Quantitative nuclear texture features analysis confirms WHO classification 2004 for lung carcinomas[J].Modern Pathology，2006，19（3）：453-459.
[47] NAKAZATO Y，MINAMI Y，KOBAYASHI H，et al.Nuclear grading of primary pulmonary adenocarcinomas：correlation between nulear size and prognosis[J].Cancer，2010，116（8）：2011-2019.
[48] 李艳，闫屹，张毛为，等.Ki-67表达与肺腺癌病理亚型及预后的关系[J].徐州医科大学学报，2022，42（4）：241-247.
LI Y，YAN Y，ZHANG M W，et al.Relationship between Ki-67 expression and pathological subtypes and prognosis of lung adenocarcinoma[J].Journal of Xuzhou Medical University，2022，42（4）：241-247.
[49] KRIEGSMANN M，CASADONTE R，KRIEGSMANN J，et al.Reliable entity subtyping in non-small cell lung cancer by maldi imaging mass spectrometry on formalin-fixed paraffin-embedded tissue specimens[J].Molecular & Cellular Proteomics，2016，15（10）：3081-3089.
[50] NEUMANN J M，FREITAG H，HRATMANN J S，et al.Subtyping non-small cell lung cancer by histology-guided spatial metabolomics[J].Journal of Cancer Research and Clinical Oncology，2022，148（2）：351-360.
[51] GRO?ERUESCHKAMP F，KALLENBACH-THIELTGE A，BEHRENS T，et al.Marker-free automated histopathological annotation of lung tumour subtypes by FTIR imaging[J].The Analyst，2015，140：2114-2120.
[52] YU K H，ZHANG C，BERRY G J，et al.Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features[J].Nature Communications，2016，7：12474.
[53] CARRILLO-PEREZ F，MORALES J C，CASTILLO-SECILLA D，et al.Non-small-cell lung cancer classification via RNA-Seq and histology imaging probability fusion[J].BMC Bioinformatics，2021，22（1）：1-19.
[54] NISHIO M，NISHIO M，JIMBO N，et al.Homology-based image processing for automatic classification of histopathological images of lung tissue[J].Cancers，2021，13（6）：1192.
[55] ECHLE A，RINDTORFF N T，BRINKER T J，et al.Deep learning in cancer pathology：a new generation of clinical biomarkers[J].British Journal of Cancer，2021，124（4）：686-696.
[56] WEI J W，TAFE L J，LINNIK Y A，et al.Pathologist-level classification of histologic patterns on resected lung adenocarcinoma slides with deep neural networks[J].Scientific Reports，2019，9：3358.
[57] GERTYCH A，SWIDERSKA-CHADAJ Z，MA Z，et al.Convolutional neural networks can accurately distinguish four histologic growth patterns of lung adenocarcinoma in digital slides[J].Scientific Reports，2019，9：1483.
[58] ANTONIO V，NAOAKI O，AKIRA S，et al.Classification of lung adenocarcinoma transcriptome subtypes from pathological images using deepconvolutional networks[J].International Journal of Computer Assisted Radiology and Surgery，2018，13（12）：1905-1913.
[59] SADHWANI A，CHANG H W，BEHROOZ A，et al.Comparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images[J].Scientific Reports，2021，11（1）：16605.
[60] 蔡阳阳.多原发性肺癌的基因组学研究[D].长春：吉林大学.
CAI Y Y.Genomics study of multiple primary lung cancer[D].Changchun：Jilin University，2020.
[61] ZHANG Y，ZHANG L，LI R，et al.Genetic variations in cancer-related significantly mutated genes and lung cancer susceptibility[J].Annals of Oncology，2017，7：1625-1630.
[62] JIANG J H，GAO J，CHEN C Y，et al.Circulating tumor cell methylation profiles reveal the classification and evolution of non-small cell lung cancer[J].Translational Lung Cancer Research，2022，11（2），224-237.
[63] HU F Y，ZHOU Y，WANG Q，et al.Gene expression classification of lung adenocarcinoma into molecular subtypes[J].IEEE/ACM Transactions on Computational Biology and Bioinformatics，2019，17（4）：1187-1197.
[64] XU D，ZHANG J L，XU H X，et al.Multi-scale supervised clustering-based feature selection for tumor classification and identification of biomarkers and targets on genomic data[J].BMC Genomics，2020，21（1）：650.
[65] DHAHBI J M，CHEN J W.Lung adenocarcinoma and lung squamous cell carcinoma cancer classification，biomarker identification，and gene-expression analysis using overlapping feature selection methods[J].Scientific Reports，2021，11（1）：13323.
[66] 周兵，王洁，彭丽姿，等.浸润性肺腺癌组织学亚型与驱动基因状态的相关性研究[J].临床医药实践，2021，30（11）：810-816.
ZHOU B，WANG J，PENG L Z，et al.Correlation between histological subtypes and driver gene status in invasive lung adenocarcinoma[J].Clinical Medicine Practice，2021，30（11）：810-816.
[67] ZHOU M，LEUNG A，ECHEGARAY S，et al.Non-small cell lung cancer radiogenomics map identifies relationships between molecular and imaging phenotypes with prognostic implications[J].Radiology，2018，286（1）：307-315.
[68] CHI W W，CHAUDHRY A.Radiogenomics of lung cancer[J].Journal of Thoracic Disease，2020，12（9）：5104-5109.
[69] SHAN Q Q，ZHANG Y F，LIANG Z A.Clustering analysis and prognostic signature of lung adenocarcinoma based on the tumor microenvironment[J].Scientific Reports，2022，12（1）：12059.
[70] KRIZHEVSKY A，SUTSKEVER I，HINTONG E.Image-Net classification with deep convolutional neural networks[C]//Communications of the ACM，2017，60（6）：84-90.
[71] ZISSERMAN A，SIMONYAN K.Very deep convolutional networks for large-scale image recognition[C]//2015 IEEE Conferenceon Computer Vision and Pattern Recognition，2015.
[72] HE K M，ZHANG X Y，REN S Q，et al.Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition，2016：770-778.
[73] HUANG G，LIU Z，WEINBERGER K Q，et al.Densely connected convolutional networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition，2017：2261-2269.
[74] BAHDANAU D，CHO K，BENGIO Y.Neural machine translation by jointly learning to align and translate[J].arXiv：1409.0473，2014.