DOI QR코드

DOI QR Code

컴퓨터 단층 촬영 영상에서의 전이성 척추 종양의 정량적 분류를 위한 라디오믹스 기반의 머신러닝 기법

Radiomics-based Machine Learning Approach for Quantitative Classification of Spinal Metastases in Computed Tomography

  • 이은우 (가천대학교 보건과학대학 의용생체공학과) ;
  • 임상헌 (가천대학교 의과대학 의공학과교실) ;
  • 전지수 (가천대학교 의과대학 의공학과교실) ;
  • 강혜원 (가천대학교 보건과학대학 의용생체공학과) ;
  • 김영재 (가천대학교 보건과학대학 의용생체공학과) ;
  • 전지영 (가천대학교 길병원 영상의학과) ;
  • 김광기 (가천대학교 보건과학대학 의용생체공학과)
  • Lee, Eun Woo (Department of Biomedical Engineering, College of Health Science, Gachon University) ;
  • Lim, Sang Heon (Department of Biomedical Engineering, Gil Medical Center, College of Medicine, Gachon University) ;
  • Jeon, Ji Soo (Department of Biomedical Engineering, Gil Medical Center, College of Medicine, Gachon University) ;
  • Kang, Hye Won (Department of Biomedical Engineering, College of Health Science, Gachon University) ;
  • Kim, Young Jae (Department of Biomedical Engineering, College of Health Science, Gachon University) ;
  • Jeon, Ji Young (Department of Radiology, Gil Medical Center, Gachon University College of Medicine) ;
  • Kim, Kwang Gi (Department of Biomedical Engineering, College of Health Science, Gachon University)
  • 투고 : 2021.03.11
  • 심사 : 2021.06.18
  • 발행 : 2021.06.30

초록

Currently, the naked eyes-based diagnosis of bone metastases on CT images relies on qualitative assessment. For this reason, there is a great need for a state-of-the-art approach that can assess and follow-up the bone metastases with quantitative biomarker. Radiomics can be used as a biomarker for objective lesion assessment by extracting quantitative numerical values from digital medical images. In this study, therefore, we evaluated the clinical applicability of non-invasive and objective bone metastases computer-aided diagnosis using radiomics-based biomarkers in CT. We employed a total of 21 approaches consist of three-classifiers and seven-feature selection methods to predict bone metastases and select biomarkers. We extracted three-dimensional features from the CT that three groups consisted of osteoblastic, osteolytic, and normal-healthy vertebral bodies. For evaluation, we compared the prediction results of the classifiers with the medical staff's diagnosis results. As a result of the three-class-classification performance evaluation, we demonstrated that the combination of the random forest classifier and the sequential backward selection feature selection approach reached AUC of 0.74 on average. Moreover, we confirmed that 90-percentile, kurtosis, and energy were the features that contributed high in the classification of bone metastases in this approach. We expect that selected quantitative features will be helpful as biomarkers in improving the patient's survival and quality of life.

키워드

과제정보

This research was supported by the Gil Medical Center (FRD2019-11-02), and by the GRRC program of Gyeonggi Province (No. GRRC Gachon 2020-B01).

참고문헌

  1. Mastro AM, Gay CV, Welch DR. The skeleton as a unique environment for breast cancer cells. Clinical & Experimental Metastasis. 2003;20(3):275-84. https://doi.org/10.1023/A:1022995403081
  2. David Roodman G, Silbermann R. Mechanisms of osteolytic and osteoblastic skeletal lesions. Bonekey Rep. 2015;4:753.
  3. Shin SO, Kim SK, Kim MS. Pallative effect of radiation therapy in management of symptomatic osseous metastases. Yeungnam University Journal of Medicine. 1992;9(1):102. https://doi.org/10.12701/yujm.1992.9.1.102
  4. Sohn SK et al. Collective Review of Cases of Spinal Metastases. The Korean Orthopaedic. 1988;23(4):1087-96. https://doi.org/10.4055/jkoa.1988.23.4.1087
  5. Batson OB, MA. The Fucntion of the Vertearal Veins and Their Role in the Spread of Metastases. Annals of Sugery. 1940;112(1):138-49. https://doi.org/10.1097/00000658-194007000-00016
  6. Liaw C-C, et al. Hepatocellular Carcinoma Presenting as Bone Metastasis. Cancer. 1989; 64(8):1753-57. https://doi.org/10.1002/1097-0142(19891015)64:8<1753::AID-CNCR2820640833>3.0.CO;2-N
  7. Park J-M. Interventional treatments for cancer pain due to bone metastasis. Anesthesia and Pain Medicine. 2015;10(3):149-64. https://doi.org/10.17085/apm.2015.10.3.149
  8. Koom WS, et al. Radiation Therapy for Bone Metastasis from Hepatocellular Carcinoma. Clinical and Molecular Hepatology. 2002;8(3):304-11.
  9. Heindel W. et al. The diagnostic imaging of bone metastases. Deutsches Arzteblatt international. 2014;111(44):741-47.
  10. HM, B.-C., J.-Z. O, M. JD. Diagnosis and Treatment Options of Spinal Metastases. Rev Invest Clin. 2015;67(3):140-57.
  11. Wang Z, et al. Exploiting machine learning for predicting skeletal-related events in cancer patients with bone metastases. Oncotarget. 2015;11(7):12612-22.
  12. Eun PJ, KH Sung Radiomics as a Quantitative Imaging Biomarker: Practical Considerations and the Current Standpoint in Neuro-oncologic Studies. Nuclear medicine and molecular imaging : NMMI. 2018;52(2):99-108. https://doi.org/10.1007/s13139-017-0512-7
  13. Ahn SJ, et al. Contrast-enhanced T1-weighted image radiomics of brain metastases may predict EGFR mutation status in primary lung cancer. Scientific Reports. 2020;10(1):8905. https://doi.org/10.1038/s41598-020-65470-7
  14. Kim KH. Non-small cell lung cancer recurrence prediction model using deep learning-based radiomics. Conf Proc 대한 기계학회 춘추학술대회, 2020. p. 53.
  15. Lee G, Bak SH, Lee HY. CT Radiomics in Thoracic Oncology: Technique and Clinical Applications. Nuclear Medicine and Molecular Imaging. 2018;52(2):91-8. https://doi.org/10.1007/s13139-017-0506-5
  16. A F., et al. 3D Slicer as an Image Computing Platform for the Quantitative Imaging Network. Magn Reson Imaging. 2012; 30(9):1323-41. https://doi.org/10.1016/j.mri.2012.05.001
  17. Yang Y, et al. Optimizing Texture Retrieving Model for Multimodal MR Image-Based Support Vector Machine for Classifying Glioma. J Magn Reson Imaging. 2019;49(5):1263-74. https://doi.org/10.1002/jmri.26524
  18. McKay C, Fujinaga I, Depalle, P. jAudio: A feature extraction library. in Proceedings of the International Conference on Music Information Retrieval. 2005.
  19. Drotar P, Gazda J, Smekal Z. An experimental comparison of feature selection methods on two-class biomedical datasets. Computers in Biology and Medicine. 2015;66:1-10. https://doi.org/10.1016/j.compbiomed.2015.08.010
  20. Kuffner R, et al. Inferring gene regulatory networks by ANOVA. Bioinformatics. 2012;28(10):1376-82. https://doi.org/10.1093/bioinformatics/bts143
  21. Chandrashekar G, Sahin F. A survey on feature selection methods. Computers & Electrical Engineering. 2014;40(1):16-28. https://doi.org/10.1016/j.compeleceng.2013.11.024
  22. Yan K, Zhang D. Feature selection and analysis on correlated gas sensor data with recursive feature elimination. Sensors and Actuators B: Chemical. 2015;212:353-63. https://doi.org/10.1016/j.snb.2015.02.025
  23. Lu M. Embedded feature selection accounting for unknown data heterogeneity. Expert Systems with Applications. 2019;119:350-61. https://doi.org/10.1016/j.eswa.2018.11.006
  24. Altmann A., et al. Permutation importance: a corrected feature importance measure. Bioinformatics. 2010;26(10):1340-47. https://doi.org/10.1093/bioinformatics/btq134
  25. Verma C, Illes Z, Sttofova V. Real-time classification of national and international students for ICT and mobile technology: an experimental study on Indian and Hungarian University. Journal of Physics: Conference Series. 2020;1432:12091. https://doi.org/10.1088/1742-6596/1432/1/012091
  26. Fluss R, Faraggi D, Reiser B. Estimation of the Youden Index and its associated cutoff point. Biom J. 2005;47(4):458-72. https://doi.org/10.1002/bimj.200410135
  27. Armstrong RA, Slade SV, Eperjesi F. An introduction to analysis of variance (ANOVA) with special reference to data from clinical experiments in optometry. Ophthalmic and Physiological Optics. 2000;20(3):235-41. https://doi.org/10.1016/S0275-5408(99)00064-2
  28. Shin DS, Ryu SM, Park CH. The Diagnostic Strategy for Malignant Bone Tumors. The Korean Orthopaedic. 2015; 50(6):429-37. https://doi.org/10.4055/jkoa.2015.50.6.429