-
肿瘤的精确诊断以及对治疗反应、不良反应和预后的准确预测是实现肿瘤精准放疗的前提。医学影像技术如超声、CT、PET/CT和MRI等以非侵入的方式获取人体组织的影像,为临床实践提供了丰富的影像资料,从而在肿瘤的个体化治疗过程中扮演着重要的角色。
以往,个体化治疗主要利用基因组学和蛋白组学技术,通过评估活检或外科手术获取的小部分肿瘤标本来分析病灶的分子生物学特征。然而,小部分肿瘤组织并不能代表肿瘤病灶的整体特点,且肿瘤的发展存在时间和空间的异质性,使其难以进行重复活检[1]。而影像学检查可对原发灶和转移病灶进行可重复的无损检查,为肿瘤进展监测和疗效评估提供了更加全面的观察视角。
目前,临床中的影像分析局限于影像科医师对病变区域的主观判断,如分析病灶的形态、位置、均匀性、强化模式以及对周围组织的浸润,并以定性的方式给出结论,进而实现临床诊断。该过程依赖阅片医师的临床知识储备和个人经验,带有一定的主观性和局限性。尽管肿瘤形态学模式的改变可以反映治疗效果,但却不能在治疗前有效预测患者的治疗反应和预后生存,因而需要更全面、深入地发掘医学影像中未被充分利用的有效信息。
事实上,高分辨率的医学影像设备除了显示常规的形态学征象外,还包含了人眼视觉分析无法捕捉到的更深层次的信息。影像组学方法作为一种非侵入性的图像分析技术,采用特征提取算法量化肿瘤影像ROI的像素灰度与空间分布的关系,发掘影像特征与临床数据间的深层关系,解码隐含在医学影像背后的由基因、细胞、生化和遗传变异等多种因素共同决定的宏观影像信息,并定量且客观地将其呈现,从而辅助指导放疗疗效和预后的评估,对原有的个体化精准放疗理论进行补充,为放疗临床实践提供新的解决方案。
本综述介绍影像组学的兴起和发展,梳理基于特征工程的影像组学和基于深度学习的影像组学在肿瘤放疗领域的研究进展,并总结研究动向,讨论影像组学与肿瘤放疗相结合的前景和面临的挑战。
影像组学与深度学习在肿瘤放疗中的研究进展
Research progress of radiomics and deep learning in tumor radiotherapy
-
摘要: 影像组学作为一种非侵入性的图像分析方法,能够深度发掘隐藏在医学影像背后的临床信息。深度学习技术的发展将影像组学研究提升到了新的高度,大量研究结果证实了其在肿瘤放疗中的应用价值。笔者从影像组学的研究背景出发,就其在肿瘤放疗中的研究进展进行综述。Abstract: As a non-invasive method of image analysis, radiomics can deeply explore the clinical information hidden behind medical images. The development of deep learning technology has promoted radiomics research to a new level, and numerous studies have confirmed its application value in tumor radiotherapy. Based on the research background of radiomics, this paper reviews its research progress in tumor radiotherapy.
-
Key words:
- Deep learning /
- Radiotherapy /
- Diagnostic imaging /
- Artificial intelligence /
- Radiomics
-
[1] Marusyk A, Almendro V, Polyak K. Intra-tumour heterogeneity: a looking glass for cancer?[J]. Nat Rev Cancer, 2012, 12(5): 323−334. DOI: 10.1038/nrc3261. [2] Lambin P, Rios-Velazquez E, Leijenaar R, et al. Radiomics: extracting more information from medical images using advanced feature analysis[J]. Eur J Cancer, 2012, 48(4): 441−446. DOI: 10.1016/j.ejca.2011.11.036. [3] Kumar V, Gu YH, Basu S, et al. Radiomics: the process and the challenges[J]. Magn Reson Imaging, 2012, 30(9): 1234−1248. DOI: 10.1016/j.mri.2012.06.010. [4] Doroshow JH, Kummar S. Translational research in oncology — 10 years of progress and future prospects[J]. Nat Rev Clin Oncol, 2014, 11(11): 649−662. DOI: 10.1038/nrclinonc.2014.158. [5] Lambin P, Leijenaar RTH, Deist TM, et al. Radiomics: the bridge between medical imaging and personalized medicine[J]. Nat Rev Clin Oncol, 2017, 14(12): 749−762. DOI: 10.1038/nrclinonc.2017.141. [6] 闫梦梦, 王卫东, 郎锦义. 影像组学技术及其在肿瘤精准放疗中的应用[J]. 肿瘤预防与治疗, 2018, 31(5): 364−368. DOI: 10.3969/j.issn.1674-0904.2018.05.011.
Yan MM, Wang WD, Lang JY. Radiomics and its application in precision radiotherapy[J]. J Cancer Control Treat, 2018, 31(5): 364−368. DOI: 10.3969/j.issn.1674-0904.2018.05.011.[7] 张佳佳, 樊鑫, 秦珊珊, 等. 基于深度学习的人工智能在肿瘤诊断中的应用进展[J]. 国际放射医学核医学杂志, 2020, 44(1): 11−15. DOI: 10.3760/cma.j.issn.1673-4114.2020.01.004.
Zhang JJ, Fan X, Qin SS, et al. Advances in the application of artificial intelligence in cancer diagnosis and treatment[J]. Int Radiat Med Nucl Med, 2020, 44(1): 11−15. DOI: 10.3760/cma.j.issn.1673-4114.2020.01.004.[8] Kamiya A, Murayama S, Kamiya H, et al. Kurtosis and skewness assessments of solid lung nodule density histograms: differentiating malignant from benign nodules on CT[J]. Jpn J Radiol, 2014, 32(1): 14−21. DOI: 10.1007/s11604-013-0264-y. [9] Li ZJ, Mao Y, Li HS, et al. Differentiating brain metastases from different pathological types of lung cancers using texture analysis of T1 postcontrast MR[J]. Magn Reson Med, 2016, 76(5): 1410−1419. DOI: 10.1002/mrm.26029. [10] Li ZJ, Mao Y, Huang W, et al. Texture-based classification of different single liver lesion based on SPAIR T2W MRI images[J]. BMC Med Imaging, 2017, 17(1): 42. DOI: 10.1186/s12880-017-0212-x. [11] Liang CS, Huang YQ, He L, et al. The development and validation of a CT-based radiomics signature for the preoperative discrimination of stage Ⅰ-Ⅱ and stage Ⅲ-Ⅳ colorectal cancer[J/OL]. Oncotarget, 2016, 7(21): 31401−31412[2020-04-01]. https://www.oncotarget.com/article/8919/text. DOI: 10.18632/oncotarget.8919. [12] Liu Y, Kim J, Balagurunathan Y, et al. Radiomic features are associated with EGFR mutation status in lung adenocarcinomas[J]. Clin Lung Cancer, 2016, 17(5): 441−448. DOI: 10.1016/j.cllc.2016.02.001. [13] Li ZJ, Wang YY, Yu JH, et al. Deep learning based radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma[J/OL]. Sci Rep, 2017, 7(1): 5467[2020-04-01]. https://www.nature.com/articles/s41598-017-05848-2. DOI: 10.1038/s41598-017-05848-2. [14] Han L, Kamdar MR. MRI to MGMT: predicting methylation status in glioblastoma patients using convolutional recurrent neural networks[J]. Pac Symp on Biocomput, 2018, 23: 331−342. DOI: 10.1142/9789813235533_0031. [15] Thakkar JP, Dolecek TA, Horbinski C, et al. Epidemiologic and molecular prognostic review of glioblastoma[J]. Cancer Epidemiol Biomarkers Prev, 2014, 23(10): 1985−1996. DOI: 10.1158/1055-9965.EPI-14-0275. [16] Feng M, Demiroz C, Vineberg K, et al. Intra-observer variability of organs at risk for head and neck cancer: geometric and dosimetric consequences[J]. Int J Radiat Oncol Biol Phys, 2010, 78(3 Suppl): S444−445. DOI: 10.1016/j.ijrobp.2010.07.1044. [17] Hardcastle N, Tomé WA, Cannon DM, et al. A multi-institution evaluation of deformable image registration algorithms for automatic organ delineation in adaptive head and neck radiotherapy[J/OL]. Radiat Oncol, 2012, 7(1): 90[2020-04-01]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3405479. DOI: 10.1186/1748-717X-7-90. [18] Li XA, Tai A, Arthur DW, et al. Variability of target and normal structure delineation for breast cancer radiotherapy: an RTOG multi-institutional and multiobserver study[J]. Int J Radiat Oncol Biol Phys, 2009, 73(3): 944−951. DOI: 10.1016/j.ijrobp.2008.10.034. [19] 周正东, 李剑波, 辛润超, 等. 基于带孔U-net神经网络的肺癌危及器官并行分割方法[J]. 东南大学学报: 自然科学版., 2019, 49(2): 231−236. DOI: 10.3969/j.issn.1001-0505.2019.02.005.
Zhou ZD, Li JB, Xin RC, et al. Parallel segmentation method for organs at risk in lung cancer based on dilated U-net neural network[J]. J Southeast Univ (Nat Sci Ed), 2019, 49(2): 231−236. DOI: 10.3969/j.issn.1001-0505.2019.02.005.[20] Kazemifar S, Balagopal A, Nguyen D, et al. Segmentation of the prostate and organs at risk in male pelvic CT images using deep learning[J/OL]. Biomed Phys Eng Express, 2018, 4(5): 055003[2020-04-01]. https://iopscience.iop.org/article/10.1088/2057-1976/aad100. DOI: 10.1088/2057-1976/aad100. [21] Liang SJ, Tang F, Huang X, et al. Deep-learning-based detection and segmentation of organs at risk in nasopharyngeal carcinoma computed tomographic images for radiotherapy planning[J]. Eur Radiol, 2019, 29(4): 1961−1967. DOI: 10.1007/s00330-018-5748-9. [22] Trebeschi S, van Griethuysen JJM, Lambregts DMJ, et al. Deep learning for fully-automated localization and segmentation of rectal cancer on multiparametric MR[J/OL]. Sci Rep, 2017, 7(1): 5301[2020-04-01]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5509680. DOI: 10.1038/s41598-017-05728-9. [23] Cardenas CE, McCarroll RE, Court LE, et al. Deep learning algorithm for auto-delineation of high-risk oropharyngeal clinical target volumes with built-in dice similarity coefficient parameter optimization function[J]. Int J Radiat Oncol Biol Phys, 2018, 101(2): 468−478. DOI: 10.1016/j.ijrobp.2018.01.114. [24] Zhuge Y, Krauze AV, Ning H, et al. Brain tumor segmentation using holistically nested neural networks in MRI images[J]. Med Phys, 2017, 44(10): 5234−5243. DOI: 10.1002/mp.12481. [25] Men K, Chen XY, Zhang Y, et al. Deep deconvolutional neural network for target segmentation of nasopharyngeal cancer in planning computed tomography images[J/OL]. Front Oncol, 2017, 7: 315[2020-04-01]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5770734. DOI: 10.3389/fonc.2017.00315. [26] Men K, Zhang T, Chen XY, et al. Fully automatic and robust segmentation of the clinical target volume for radiotherapy of breast cancer using big data and deep learning[J]. Phys Med, 2018, 50: 13−9. DOI: 10.1016/j.ejmp.2018.05.006. [27] Boon IS, Au Yong TPT, Boon C. Assessing the role of artificial intelligence (AI) in clinical oncology: utility of machine learning in radiotherapy target volume delineation[J/OL]. Medicines (Basel), 2018, 5(4): 131[2020-04-01]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6313566. DOI: 10.3390/medicines5040131. [28] Meyer P, Noblet V, Mazzara C, et al. Survey on deep learning for radiotherapy[J]. Comput Biol Med, 2018, 98: 126−146. DOI: 10.1016/j.compbiomed.2018.05.018. [29] Nguyen D, Long T, Jia X, et al. A feasibility study for predicting optimal radiation therapy dose distributions of prostate cancer patients from patient anatomy using deep learning[J/OL]. Sci Rep, 2019, 9(1): 1076[2020-04-01]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6355802. DOI: 10.1038/s41598-018-37741-x. [30] Chen XY, Men K, Li YX, et al. A feasibility study on an automated method to generate patient-specific dose distributions for radiotherapy using deep learning[J]. Med Phys, 2019, 46(1): 56−64. DOI: 10.1002/mp.13262. [31] Fan JW, Wang JZ, Chen Z, et al. Automatic treatment planning based on three-dimensional dose distribution predicted from deep learning technique[J]. Med Phys, 2019, 46(1): 370−381. DOI: 10.1002/mp.13271. [32] Babier A, Mahmood R, McNiven AL, et al. Knowledge-based automated planning with three-dimensional generative adversarial networks[J]. Med Phys, 2020, 47(2): 297−306. DOI: 10.1002/mp.13896. [33] Liu J, Mao Y, Li ZJ, et al. Use of texture analysis based on contrast-enhanced MRI to predict treatment response to chemoradiotherapy in nasopharyngeal carcinoma[J]. J Magn Reson Imaging, 2016, 44(2): 445−455. DOI: 10.1002/jmri.25156. [34] Cusumano D, Dinapoli N, Boldrini L, et al. Fractal-based radiomic approach to predict complete pathological response after chemo-radiotherapy in rectal cancer[J]. Radiol Med, 2018, 123(4): 286−295. DOI: 10.1007/s11547-017-0838-3. [35] Bibault JE, Giraud P, Housset M, et al. Deep learning and radiomics predict complete response after neo-adjuvant chemoradiation for locally advanced rectal cancer[J/OL]. Sci Rep, 2018, 8(1): 1−8[2020-04-01]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6105676. DOI: 10.1038/s41598-018-30657-6. [36] Coroller TP, Agrawal V, Huynh E, et al. Radiomic-based pathological response prediction from primary tumors and lymph nodes in NSCLC[J]. J Thorac Oncol, 2017, 12(3): 467−476. DOI: 10.1016/j.jtho.2016.11.2226. [37] Moran A, Daly ME, Yip SS, et al. Radiomics-based assessment of radiation-induced lung injury after stereotactic body radiotherapy[J]. Clin Lung Cancer, 2017, 18(6): e425−e431. DOI: 10.1016/j.cllc.2017.05.014. [38] Ibragimov B, Toesca D, Chang D, et al. Development of deep neural network for individualized hepatobiliary toxicity prediction after liver SBRT[J]. Med Phys, 2018, 45(10): 4763−4774. DOI: 10.1002/mp.13122. [39] Zhen X, Chen JW, Zhong ZC, et al. Deep convolutional neural network with transfer learning for rectum toxicity prediction in cervical cancer radiotherapy: a feasibility study[J]. Phys Med Biol, 2017, 62(21): 8246−8263. DOI: 10.1088/1361-6560/aa8d09. [40] Zhang B, Tian J, Dong D, et al. Radiomics features of multiparametric MRI as novel prognostic factors in advanced nasopharyngeal carcinoma[J]. Clin Cancer Res, 2017, 23(15): 4259−4269. DOI: 10.1158/1078-0432.CCR-16-2910. [41] van Timmeren JE, Leijenaar RTH, van Elmpt W, et al. Survival prediction of non-small cell lung cancer patients using radiomics analyses of cone-beam CT images[J]. Radiother Oncol, 2017, 123(3): 363−369. DOI: 10.1016/j.radonc.2017.04.016. [42] Lao JW, Chen YS, Li ZC, et al. A deep learning-based radiomics model for prediction of survival in glioblastoma multiforme[J/OL]. Sci Rep, 2017, 7(1): 10353[2020-04-01]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5583361. DOI: 10.1038/s41598-017-10649-8. [43] Shen WC, Chen SW, Wu KC, et al. Prediction of local relapse and distant metastasis in patients with definitive chemoradiotherapy-treated cervical cancer by deep learning from [18F]-fluorodeoxyglucose positron emission tomography/computed tomography[J]. Eur Radiol, 2019, 29(12): 6741−6749. DOI: 10.1007/s00330-019-06265-x. [44] Dissaux G, Visvikis D, Da-Ano R, et al. Pretreatment 18F-FDG PET/CT radiomics predict local recurrence in patients treated with stereotactic radiotherapy for early-stage non-small cell lung cancer: a multicentric study[J]. J Nucl Med, 2020, 61(6): 814−820. DOI: 10.2967/jnumed.119.228106. [45] Cao Q, Li YM, Li Z, et al. Development and validation of a radiomics signature on differentially expressed features of 18F-FDG PET to predict treatment response of concurrent chemoradiotherapy in thoracic esophagus squamous cell carcinoma[J]. Radiother Oncol, 2020, 146: 9−15. DOI: 10.1016/j.radonc.2020.01.027. [46] He L, Huang YQ, Ma ZL, et al. Effects of contrast-enhancement, reconstruction slice thickness and convolution kernel on the diagnostic performance of radiomics signature in solitary pulmonary nodule[J/OL]. Sci Rep, 2016, 6: 34921[2020-04-01]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5056507. DOI: 10.1038/srep34921. [47] Fave X, Cook M, Frederick A, et al. Preliminary investigation into sources of uncertainty in quantitative imaging features[J]. Comput Med Imaging Graph, 2015, 44: 54−61. DOI: 10.1016/j.compmedimag.2015.04.006. [48] Yang JZ, Zhang LF, Fave XJ, et al. Uncertainty analysis of quantitative imaging features extracted from contrast-enhanced CT in lung tumors[J]. Comput Med Imaging Graph, 2016, 48: 1−8. DOI: 10.1016/j.compmedimag.2015.12.001.
计量
- 文章访问数: 13362
- HTML全文浏览量: 11759
- PDF下载量: 53