Korean Journal of Radiology

  • 제21권7호
  • /
  • Pages.779-792
  • /
  • 2020
  • /
  • 1229-6929(pISSN)
  • /
  • 2005-8330(eISSN)
대한영상의학회 (The Korean Society of Radiology)
권호 표지
DOI QR코드

DOI QR Code

Radiomics in Breast Imaging from Techniques to Clinical Applications: A Review

  • Seung-Hak Lee (Department of Electrical and Computer Engineering, Sungkyunkwan University) ;
  • Hyunjin Park (Center for Neuroscience Imaging Research, Institute for Basic Science (IBS)) ;
  • Eun Sook Ko (Department of Radiology, Samsung Medical Center, Sungkyunkwan University School of Medicine)
  • 투고 : 2019.11.12
  • 심사 : 2020.02.28
  • 발행 : 2020.07.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

Recent advances in computer technology have generated a new area of research known as radiomics. Radiomics is defined as the high throughput extraction and analysis of quantitative features from imaging data. Radiomic features provide information on the gray-scale patterns, inter-pixel relationships, as well as shape and spectral properties of radiological images. Moreover, these features can be used to develop computational models that may serve as a tool for personalized diagnosis and treatment guidance. Although radiomics is becoming popular and widely used in oncology, many problems such as overfitting and reproducibility issues remain unresolved. In this review, we will outline the steps of radiomics used for oncology, specifically addressing applications for breast cancer patients and focusing on technical issues.

키워드

참고문헌

  1. Limkin EJ, Sun R, Dercle L, Zacharaki EI, Robert C, Reuze S, et al. Promises and challenges for the implementation of computational medical imaging (radiomics) in oncology. Ann Oncol 2017;28:1191-1206
  2. Khangura S, Konnyu K, Cushman R, Grimshaw J, Moher D. Evidence summaries: the evolution of a rapid review approach. Syst Rev 2012;1:10
  3. Gillies RJ, Kinahan PE, Hricak H. Radiomics: images are more than pictures, they are data. Radiology 2016;278:563-577
  4. Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, van Stiphout RG, Granton P, et al. Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer 2012;48:441-446
  5. Kumar V, Gu Y, Basu S, Berglund A, Eschrich SA, Schabath MB, et al. Radiomics: the process and the challenges. Magn Reson Imaging 2012;30:1234-1248
  6. Haralick RM, Shanmugam K, Dinstein I. Textural features for image classification. IEEE Trans Syst Man Cybern 1973;SMC-3:610-621
  7. Mandelbrot BB. The fractal geometry of nature. San Francisco: Freeman and co., 1982
  8. Amadasun M, King R. Textural features corresponding to textural properties. IEEE Trans Syst Man Cybern Syst 1989;19:1264-1274
  9. Aerts HJ, Velazquez ER, Leijenaar RT, Parmar C, Grossmann P, Carvalho S, et al. Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nat Commun 2014;5:4006
  10. Larue RT, Defraene G, De Ruysscher D, Lambin P, van Elmpt W. Quantitative radiomics studies for tissue characterization: a review of technology and methodological procedures. Br J Radiol 2017;90:20160665
  11. Aerts HJ. The potential of radiomic-based phenotyping in precision medicine: a review. JAMA Oncol 2016;2:1636-1642
  12. Peerlings J, Woodruff HC, Winfield JM, Ibrahim A, Van Beers BE, Heerschap A, et al. Stability of radiomics features in apparent diffusion coefficient maps from a multi-centre test-retest trial. Sci Rep 2019;9:4800
  13. van Timmeren JE, Leijenaar RTH, van Elmpt W, Wang J, Zhang Z, Dekker A, et al. Test-retest data for radiomics feature stability analysis: generalizable or study-specific? Tomography 2016;2:361-365
  14. Larue RTHM, van Timmeren JE, de Jong EEC, Feliciani G, Leijenaar RTH, Schreurs WMJ, et al. Influence of gray level discretization on radiomic feature stability for different CT scanners, tube currents and slice thicknesses: a comprehensive phantom study. Acta Oncol 2017;56:1544-1553
  15. Mackin D, Fave X, Zhang L, Fried D, Yang J, Taylor B, et al. Measuring computed tomography scanner variability of radiomics features. Invest Radiol 2015;50:757-765
  16. Xu Y, Hosny A, Zeleznik R, Parmar C, Coroller T, Franco I, et al. Deep learning predicts lung cancer treatment response from serial medical imaging. Clin Cancer Res 2019;25:3266-3275
  17. Huang SY, Franc BL, Harnish RJ, Liu G, Mitra D, Copeland TP, et al. Exploration of PET and MRI radiomic features for decoding breast cancer phenotypes and prognosis. NPJ Breast Cancer 2018;4:24
  18. Rios Velazquez E, Aerts HJ, Gu Y, Goldgof DB, De Ruysscher D, Dekker A, et al. A semiautomatic CT-based ensemble segmentation of lung tumors: comparison with oncologists' delineations and with the surgical specimen. Radiother Oncol 2012;105:167-173
  19. Heye T, Merkle EM, Reiner CS, Davenport MS, Horvath JJ, Feuerlein S, et al. Reproducibility of dynamic contrast-enhanced MR imaging. Part II. Comparison of intra- and interobserver variability with manual region of interest placement versus semiautomatic lesion segmentation and histogram analysis. Radiology 2013;266:812-821
  20. Grove O, Berglund AE, Schabath MB, Aerts HJ, Dekker A, Wang H, et al. Quantitative computed tomographic descriptors associate tumor shape complexity and intratumor heterogeneity with prognosis in lung adenocarcinoma. PLoS One 2015;10:e0118261
  21. Nam KJ, Park H, Ko ES, Lim Y, Cho HH, Lee JE. Radiomics signature on 3T dynamic contrast-enhanced magnetic resonance imaging for estrogen receptor-positive invasive breast cancers: preliminary results for correlation with Oncotype DX recurrence scores. Medicine (Baltimore) 2019;98:e15871
  22. Lee SH, Cho HH, Lee HY, Park H. Clinical impact of variability on CT radiomics and suggestions for suitable feature selection: a focus on lung cancer. Cancer Imaging 2019;19:54
  23. Park H, Lim Y, Ko ES, Cho HH, Lee JE, Han BK, et al. Radiomics signature on magnetic resonance imaging: association with disease-free survival in patients with invasive breast cancer. Clin Cancer Res 2018;24:4705-4714
  24. Aerts HJ, Grossmann P, Tan Y, Oxnard GR, Rizvi N, Schwartz LH, et al. Defining a radiomic response phenotype: a pilot study using targeted therapy in NSCLC. Sci Rep 2016;6:33860
  25. Parmar C, Grossmann P, Bussink J, Lambin P, Aerts HJWL. Machine learning methods for quantitative radiomic biomarkers. Sci Rep 2015;5:13087
  26. Liang C, Huang Y, He L, Chen X, Ma Z, Dong D, 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. Oncotarget 2016;7:31401-31412
  27. Chen W, Liu B, Peng S, Sun J, Qiao X. Computer-aided grading of gliomas combining automatic segmentation and radiomics. Int J Biomed Imaging 2018;2018:2512037
  28. Napel S, Mu W, Jardim-Perassi BV, Aerts HJWL, Gillies RJ. Quantitative imaging of cancer in the postgenomic era: radio(geno)mics, deep learning, and habitats. Cancer 2018;124:4633-4649
  29. Cho H, Lee S, Kim J, Park H. Classification of the glioma grading using radiomics analysis. PeerJ 2018;6:e5982
  30. Liu Y, Zhang Y, Cheng R, Liu S, Qu F, Yin X, et al. Radiomics analysis of apparent diffusion coefficient in cervical cancer: a preliminary study on histological grade evaluation. J Magn Reson Imaging 2019;49:280-290
  31. Coroller TP, Agrawal V, Narayan V, Hou Y, Grossmann P, Lee SW, et al. Radiomic phenotype features predict pathological response in non-small cell lung cancer. Radiother Oncol 2016;119:480-486
  32. Clark K, Vendt B, Smith K, Freymann J, Kirby J, Koppel P, et al. The Cancer Imaging Archive (TCIA): maintaining and operating a public information repository. J Digit Imaging 2013;26:1045-1057
  33. Traverso A, Wee L, Dekker A, Gillies R. Repeatability and reproducibility of radiomic features: a systematic review. Int J Radiat Oncol Biol Phys 2018;102:1143-1158
  34. Lambin P, Leijenaar RTH, Deist TM, Peerlings J, de Jong EEC, van Timmeren J, et al. Radiomics: the bridge between medical imaging and personalized medicine. Nat Rev Clin Oncol 2017;14:749-762
  35. Zhao B, Tan Y, Tsai WY, Qi J, Xie C, Lu L, et al. Reproducibility of radiomics for deciphering tumor phenotype with imaging. Sci Rep 2016;6:23428
  36. Balagurunathan Y, Gu Y, Wang H, Kumar V, Grove O, Hawkins S, et al. Reproducibility and prognosis of quantitative features extracted from CT images. Transl Oncol 2014;7:72-87
  37. Balagurunathan Y, Kumar V, Gu Y, Kim J, Wang H, Liu Y, et al. Test-retest reproducibility analysis of lung CT image features. J Digit Imaging 2014;27:805-823
  38. Tixier F, Hatt M, Le Rest CC, Le Pogam A, Corcos L, Visvikis D. Reproducibility of tumor uptake heterogeneity characterization through textural feature analysis in 18F-FDG PET. J Nucl Med 2012;53:693-700
  39. Berenguer R, Pastor-Juan MDR, Canales-Vazquez J, Castro-Garcia M, Villas MV, Mansilla Legorburo F, et al. Radiomics of CT features may be nonreproducible and redundant: influence of CT acquisition parameters. Radiology 2018;288:407-415
  40. Park JE, Park SY, Kim HJ, Kim HS. Reproducibility and generalizability in radiomics modeling: possible strategies in radiologic and statistical perspectives. Korean J Radiol 2019;20:1124-1137
  41. Leijenaar RT, Nalbantov G, Carvalho S, van Elmpt WJ, Troost EG, Boellaard R, et al. The effect of SUV discretization in quantitative FDG-PET radiomics: the need for standardized methodology in tumor texture analysis. Sci Rep 2015;5:11075
  42. Fedorov A, Beichel R, Kalpathy-Cramer J, Finet J, Fillion-Robin JC, Pujol S, et al. 3D Slicer as an image computing platform for the Quantitative Imaging Network. Magn Reson Imaging 2012;30:1323-1341
  43. Bazin PL, Cuzzocreo JL, Yassa MA, Gandler W, McAuliffe MJ, Bassett SS, et al. Volumetric neuroimage analysis extensions for the MIPAV software package. J Neurosci Methods 2007;165:111-121
  44. Ger RB, Cardenas CE, Anderson BM, Yang J, Mackin DS, Zhang L, et al. Guidelines and experience using Imaging Biomarker Explorer (IBEX) for radiomics. J Vis Exp 2018;(131):e57132
  45. van Griethuysen JJM, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V, et al. Computational radiomics system to decode the radiographic phenotype. Cancer Res 2017;77:e104-e107
  46. Wilson R, Devaraj A. Radiomics of pulmonary nodules and lung cancer. Transl Lung Cancer Res 2017;6:86-91
  47. Huang Y, Liu Z, He L, Chen X, Pan D, Ma Z, et al. Radiomics signature: a potential biomarker for the prediction of disease-free survival in early-stage (I or II) non-small cell lung cancer. Radiology 2016;281:947-957
  48. Rahbar H, McDonald ES, Lee JM, Partridge SC, Lee CI. How can advanced imaging be used to mitigate potential breast cancer overdiagnosis? Acad Radiol 2016;23:768-773
  49. Bickelhaupt S, Jaeger PF, Laun FB, Lederer W, Daniel H, Kuder TA, et al. Radiomics based on adapted diffusion kurtosis imaging helps to clarify most mammographic findings suspicious for cancer. Radiology 2018;287:761-770
  50. Nie K, Chen JH, Yu HJ, Chu Y, Nalcioglu O, Su MY. Quantitative analysis of lesion morphology and texture features for diagnostic prediction in breast MRI. Acad Radiol 2008;15:1513-1525
  51. Wang TC, Huang YH, Huang CS, Chen JH, Huang GY, Chang YC, et al. Computer-aided diagnosis of breast DCE-MRI using pharmacokinetic model and 3-D morphology analysis. Magn Reson Imaging 2014;32:197-205
  52. Cai H, Peng Y, Ou C, Chen M, Li L. Diagnosis of breast masses from dynamic contrast-enhanced and diffusion-weighted MR: a machine learning approach. PLoS One 2014;9:e87387
  53. Parekh VS, Jacobs MA. Integrated radiomic framework for breast cancer and tumor biology using advanced machine learning and multiparametric MRI. NPJ Breast Cancer 2017;3:43
  54. Garra BS, Krasner BH, Horii SC, Ascher S, Mun SK, Zeman RK. Improving the distinction between benign and malignant breast lesions: the value of sonographic texture analysis. Ultrason Imaging 1993;15:267-285
  55. Luo WQ, Huang QX, Huang XW, Hu HT, Zeng FQ, Wang W. Predicting breast cancer in breast imaging reporting and data system (BI-RADS) ultrasound category 4 or 5 lesions: a nomogram combining radiomics and BI-RADS. Sci Rep 2019;9:11921
  56. Zhang Q, Xiao Y, Suo J, Shi J, Yu J, Guo Y, et al. Sonoelastomics for breast tumor classification: a radiomics approach with clustering-based feature selection on sonoelastography. Ultrasound Med Biol 2017;43:1058-1069
  57. Drukker K, Giger ML, Joe BN, Kerlikowske K, Greenwood H, Drukteinis JS, et al. Combined benefit of quantitative three-compartment breast image analysis and mammography radiomics in the classification of breast masses in a clinical data set. Radiology 2019;290:621-628
  58. Li H, Mendel KR, Lan L, Sheth D, Giger ML. Digital mammography in breast cancer: additive value of radiomics of breast parenchyma. Radiology 2019;291:15-20
  59. Tagliafico AS, Valdora F, Mariscotti G, Durando M, Nori J, La Forgia D, et al. An exploratory radiomics analysis on digital breast tomosynthesis in women with mammographically negative dense breasts. Breast 2018;40:92-96
  60. Holli K, Laaperi AL, Harrison L, Luukkaala T, Toivonen T, Ryymin P, et al. Characterization of breast cancer types by texture analysis of magnetic resonance images. Acad Radiol 2010;17:135-141
  61. Waugh SA, Purdie CA, Jordan LB, Vinnicombe S, Lerski RA, Martin P, et al. Magnetic resonance imaging texture analysis classification of primary breast cancer. Eur Radiol 2016;26:322-330
  62. Li H, Zhu Y, Burnside ES, Huang E, Drukker K, Hoadley KA, et al. Quantitative MRI radiomics in the prediction of molecular classifications of breast cancer subtypes in the TCGA/TCIA data set. NPJ Breast Cancer 2016;2. pii: 16012
  63. Liang C, Cheng Z, Huang Y, He L, Chen X, Ma Z, et al. An MRI-based radiomics classifier for preoperative prediction of Ki-67 status in breast cancer. Acad Radiol 2018;25:1111-1117
  64. Marino MA, Pinker K, Leithner D, Sung J, Avendano D, Morris EA, et al. Contrast-enhanced mammography and radiomics analysis for noninvasive breast cancer characterization: initial results. Mol Imaging Biol 2019 Aug 28 [Epub]. https://doi.org/10.1007/s11307-019-01423-5
  65. Ahmed A, Gibbs P, Pickles M, Turnbull L. Texture analysis in assessment and prediction of chemotherapy response in breast cancer. J Magn Reson Imaging 2013;38:89-101
  66. Braman NM, Etesami M, Prasanna P, Dubchuk C, Gilmore H, Tiwari P, et al. Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI. Breast Cancer Res 2017;19:57
  67. Braman N, Prasanna P, Whitney J, Singh S, Beig N, Etesami M, et al. Association of peritumoral radiomics with tumor biology and pathologic response to preoperative targeted therapy for HER2 (ERBB2)-positive breast cancer. JAMA Netw Open 2019;2:e192561
  68. Liu Z, Li Z, Qu J, Zhang R, Zhou X, Li L, et al. Radiomics of multiparametric MRI for pretreatment prediction of pathologic complete response to neoadjuvant chemotherapy in breast cancer: a multicenter study. Clin Cancer Res 2019;25:3538-3547
  69. Dong Y, Feng Q, Yang W, Lu Z, Deng C, Zhang L, et al. Preoperative prediction of sentinel lymph node metastasis in breast cancer based on radiomics of T2-weighted fat-suppression and diffusion-weighted MRI. Eur Radiol 2018;28:582-591
  70. Yang J, Wang T, Yang L, Wang Y, Li H, Zhou X, et al. Preoperative prediction of axillary lymph node metastasis in breast cancer using mammography-based radiomics method. Sci Rep 2019;9:4429
  71. Yu FH, Wang JX, Ye XH, Deng J, Hang J, Yang B. Ultrasound-based radiomics nomogram: a potential biomarker to predict axillary lymph node metastasis in early-stage invasive breast cancer. Eur J Radiol 2019;119:108658
  72. Chan HM, van der Velden BHM, Loo CE, Gilhuijs KGA. Eigentumors for prediction of treatment failure in patients with early-stage breast cancer using dynamic contrast-enhanced MRI: a feasibility study. Phys Med Biol 2017;62:6467-6485
  73. Sinha S, Lucas-Quesada FA, DeBruhl ND, Sayre J, Farria D, Gorczyca DP, et al. Multifeature analysis of Gd-enhanced MR images of breast lesions. J Magn Reson Imaging 1997;7:1016-1026
  74. Gibbs P, Turnbull LW. Textural analysis of contrast-enhanced MR images of the breast. Magn Reson Med 2003;50:92-98
  75. Ertas, G, Gulcur HO, Tunaci M. Improved lesion detection in MR mammography: three-dimensional segmentation, moving voxel sampling, and normalized maximum intensity-time ratio entropy. Acad Radiol 2007;14:151-161
  76. McLaren CE, Chen WP, Nie K, Su MY. Prediction of malignant breast lesions from MRI features: a comparison of artificial neural network and logistic regression techniques. Acad Radiol 2009;16:842-851
  77. Sivaramakrishna R, Powell KA, Lieber ML, Chilcote WA, Shekhar R. Texture analysis of lesions in breast ultrasound images. Comput Med Imaging Graph 2002;26:303-307
  78. Chen DR, Chang RF, Kuo WJ, Chen MC, Huang YL. Diagnosis of breast tumors with sonographic texture analysis using wavelet transform and neural networks. Ultrasound Med Biol 2002;28:1301-1310
  79. Chen DR, Chang RF, Huang YL, Chou YH, Tiu CM, Tsai PP. Texture analysis of breast tumors on sonograms. Semin Ultrasound CT MR 2000;21:308-316
  80. Karahaliou A, Skiadopoulos S, Boniatis I, Sakellaropoulos P, Likaki E, Panayiotakis G, et al. Texture analysis of tissue surrounding microcalcifications on mammograms for breast cancer diagnosis. Br J Radiol 2007;80:648-656
  81. Wei D, Chan HP, Petrick N, Sahiner B, Helvie MA, Adler DD, et al. False-positive reduction technique for detection of masses on digital mammograms: global and local multiresolution texture analysis. Med Phys 1997;24:903-914
  82. Chan HP, Sahiner B, Lam KL, Petrick N, Helvie MA, Goodsitt MM, et al. Computerized analysis of mammographic microcalcifications in morphological and texture feature spaces. Med Phys 1998;25:2007-2019
  83. Goldhirsch A, Winer EP, Coates AS, Gelber RD, Piccart-Gebhart M, Thurlimann B, et al. Personalizing the treatment of women with early breast cancer: highlights of the St Gallen International Expert Consensus on the primary therapy of early breast cancer 2013. Ann Oncol 2013;24:2206-2223
  84. Parikh J, Selmi M, Charles-Edwards G, Glendenning J, Ganeshan B, Verma H, et al. Changes in primary breast cancer heterogeneity may augment midtreatment MR imaging assessment of response to neoadjuvant chemotherapy. Radiology 2014;272:100-112
  85. Cui X, Wang N, Zhao Y, Chen S, Li S, Xu M, et al. Preoperative prediction of axillary lymph node metastasis in breast cancer using radiomics features of DCE-MRI. Sci Rep 2019;9:2240
  86. Kim JH, Ko ES, Lim Y, Lee KS, Han BK, Ko EY, et al. Breast cancer heterogeneity: MR imaging texture analysis and survival outcomes. Radiology 2017;282:665-675
  87. Pickles MD, Lowry M, Gibbs P. Pretreatment prognostic value of dynamic contrast-enhanced magnetic resonance imaging vascular, texture, shape, and size parameters compared with traditional survival indicators obtained from locally advanced breast cancer patients. Invest Radiol 2016;51:177-185
  88. Choe J, Lee SM, Do KH, Lee G, Lee JG, Lee SM, et al. Deep learning-based image conversion of CT reconstruction kernels improves radiomics reproducibility for pulmonary nodules or masses. Radiology 2019;292:365-373
  89. Goodfellow I, Bengio Y, Courville A. Deep learning. Cambridge: MIT Press, 2016
  90. Shen G, Dwivedi K, Majima K, Horikawa T, Kamitani Y. End-to-end deep image reconstruction from human brain activity. Front Comput Neurosci 2019;13:21
  91. Shin HC, Tenenholtz NA, Rogers JK, Schwarz CG, Senjem ML, Gunter JL, et al. Medical image synthesis for data augmentation and anonymization using generative adversarial networks. International workshop on simulation and synthesis in medical imaging;2018 September 16;Granada, Spain
  92. Nie D, Trullo R, Lian J, Wang L, Petitjean C, Ruan S, et al. Medical image synthesis with deep convolutional adversarial networks. IEEE Trans Biomed Eng 2018;65:2720-2730
  93. Isin A, Direkoglu C, Sah M. Review of MRI-based brain tumor image segmentation using deep learning methods. Procedia Comput Sci 2016;102:317-324
  94. Havaei M, Davy A, Warde-Farley D, Biard A, Courville A, Bengio Y, et al. Brain tumor segmentation with deep neural networks. Med Image Anal 2017;35:18-31
  95. Afshar P, Mohammadi A, Plataniotis KN, Oikonomou A, Benali H. From handcrafted to deep-learning-based cancer radiomics: challenges and opportunities. IEEE Signal Process Mag 2019;36:132-160
  96. Gilpin LH, Bau D, Yuan BZ, Bajwa A, Specter M, Kagal L. Explaining explanations: an overview of interpretability of machine learning. 2018 IEEE 5th international conference on data science and advanced analytics (DSAA);2018 October 1-3;Turin, Italy
  97. Hosny A, Parmar C, Coroller TP, Grossmann P, Zeleznik R, Kumar A, et al. Deep learning for lung cancer prognostication: a retrospective multi-cohort radiomics study. PLoS Med 2018;15:e1002711
  98. Parmar C, Barry JD, Hosny A, Quackenbush J, Aerts HJWL. Data analysis strategies in medical imaging. Clin Cancer Res 2018;24:3492-3499
  99. Valdora F, Houssami N, Rossi F, Calabrese M, Tagliafico AS. Rapid review: radiomics and breast cancer. Breast Cancer Res Treat 2018;169:217-229