대한영상의학회지 (Journal of the Korean Society of Radiology)

  • 제81권5호
  • /
  • Pages.1164-1174
  • /
  • 2020
  • /
  • 1738-2637(pISSN)
  • /
  • 2288-2928(eISSN)
대한영상의학회 (The Korean Society of Radiology)
권호 표지
DOI QR코드

DOI QR Code

인공지능 딥러닝을 이용한 갑상선 초음파에서의 갑상선암의 재발 예측

Deep Learning in Thyroid Ultrasonography to Predict Tumor Recurrence in Thyroid Cancers

  • 길지은 (한양대학교 의과대학 영상의학과) ;
  • 김광기 (가천대학교 의과대학 의예과) ;
  • 김영재 (가천대학교 의과대학 의예과) ;
  • 구혜령 (한양대학교 의과대학 영상의학과) ;
  • 박정선 (한양대학교 의과대학 영상의학과)
  • Jieun Kil (Department of Radiology, Hanyang University College of Medicine) ;
  • Kwang Gi Kim (Department of Biomedical Engineering, College of Medicine, Gachon University) ;
  • Young Jae Kim (Department of Biomedical Engineering, College of Medicine, Gachon University) ;
  • Hye Ryoung Koo (Department of Radiology, Hanyang University College of Medicine) ;
  • Jeong Seon Park (Department of Radiology, Hanyang University College of Medicine)
  • 투고 : 2019.08.06
  • 심사 : 2019.10.08
  • 발행 : 2020.09.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

목적 수술 전 초음파 검사에서 갑상선 종양의 재발을 예측할 수 있는 심층 학습 모델을 개발하고자 한다. 대상과 방법 수술 전 초음파에서 병리학적으로 확진된 갑상선 수술을 받은 229명의 환자(남성:여성 = 42:187, 평균 연령, 49.6세)의 대표적인 초음파 이미지를 포함시켰다. 각각 대표적인 횡축 또는 종축 초음파 이미지가 선택되었다. 신경 네트워크용 Python 2.7.6 및 Keras 2.1.5, convolutional neural network을 사용한 심층 학습이 사용되었다. 재발한 환자와 재발이 없는 환자의 임상 및 조직학적 특징을 비교하였다. 그룹 간의 심층 학습 모델의 receiver operating characteristic curve 곡선 아래의 영역은 재발 갑상선암을 예측하기 위한 심층 학습 모델의 예측에 사용되었다. 결과 전체 환자 229명 중 49명이 종양 재발(21.4%)을 보였다. 종양의 크기, 다원성은 재발이 없는 군과 재발 군에서 유의한 차이가 있었다(p < 0.05). 재발성 갑상선암 예측을 위한 심층 학습 모델의 전반적인 평균 area under the curve (이하 AUC) 값은 0.9 ± 0.06이었다. 평균 AUC는 macrocarcinoma에서 0.87 ± 0.03, microcarcinoma에서 0.79 ± 0.16이었다. 결론 갑상선암의 초음파 이미지를 이용한 심층 학습 모델로 갑상선암 재발의 예측 모델 구축의 가능성을 보여주었다.

Purpose To evaluate a deep learning model to predict recurrence of thyroid tumor using preoperative ultrasonography (US). Materials and Methods We included representative images from 229 US-based patients (male:female = 42:187; mean age, 49.6 years) who had been diagnosed with thyroid cancer on preoperative US and subsequently underwent thyroid surgery. After selecting each representative transverse or longitudinal US image, we created a data set from the resulting database of 898 images after augmentation. The Python 2.7.6 and Keras 2.1.5 framework for neural networks were used for deep learning with a convolutional neural network. We compared the clinical and histological features between patients with and without recurrence. The predictive performance of the deep learning model between groups was evaluated using receiver operating characteristic (ROC) analysis, and the area under the ROC curve served as a summary of the prognostic performance of the deep learning model to predict recurrent thyroid cancer. Results Tumor recurrence was noted in 49 (21.4%) among the 229 patients. Tumor size and multifocality varied significantly between the groups with and without recurrence (p < 0.05). The overall mean area under the curve (AUC) value of the deep learning model for prediction of recurrent thyroid cancer was 0.9 ± 0.06. The mean AUC value was 0.87 ± 0.03 in macrocarcinoma and 0.79 ± 0.16 in microcarcinoma. Conclusion A deep learning model for analysis of US images of thyroid cancer showed the possibility of predicting recurrence of thyroid cancer.

키워드

과제정보

This work was supported by the research fund of Hanyang University (HY-2016). This research supported by the Gachon Univ. funding (2017-0211). Jeong Seon Park and Kwang Gi Kim are equally contributed.

참고문헌

  1. Ito Y, Kihara M, Takamura Y, Kobayashi K, Miya A, Hirokawa M, et al. Prognosis and prognostic factors of papillary thyroid carcinoma in patients under 20 years. Endocr J 2012;59:539-545 
  2. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014;64:9-29 
  3. Lim H, Devesa SS, Sosa JA, Check D, Kitahara CM. Trends in thyroid cancer incidence and mortality in the United States, 1974-2013. JAMA 2017;317:1338-1348 
  4. Stewart B, Wild CP. World Cancer Report 2014. Available at. http://publichealthwell.ie/node/725845. Published 2014. Accessed Nov 19, 2019 
  5. Baek SK, Jung KY, Kang SM, Kwon SY, Woo JS, Cho SH, et al. Clinical risk factors associated with cervical lymph node recurrence in papillary thyroid carcinoma. Thyroid 2010;20:147-152 
  6. Conzo G, Docimo G, Ruggiero R, Napolitano S, Palazzo A, Gambardella C, et al. Surgical treatment of papillary thyroid carcinoma without lymph nodal involvement. G Chir 2012;33:339-342 
  7. Cramer JD, Fu P, Harth KC, Margevicius S, Wilhelm SM. Analysis of the rising incidence of thyroid cancer using the surveillance, epidemiology and end results national cancer data registry. Surgery 2010;148:1147-1152 
  8. Hughes DT, Haymart MR, Miller BS, Gauger PG, Doherty GM. The most commonly occurring papillary thyroid cancer in the United States is now a microcarcinoma in a patient older than 45 years. Thyroid 2011;21:231-236 
  9. Jung KW, Park S, Kong HJ, Won YJ, Lee JY, Park EC, et al. Cancer statistics in Korea: incidence, mortality, survival, and prevalence in 2008. Cancer Res Treat 2011;43:1-11 
  10. Hedinger C, Williams ED, Sobin LH. Histologic typing of thyroid tumors. International histologic classification of tumors. Berlin: Springer 1988:7-68 
  11. Mazeh H, Chen H. Advances in surgical therapy for thyroid cancer. Nat Rev Endocrinol 2011;7:581-588 
  12. Mann B, Buhr HJ. Lymph node dissection in patients with differentiated thyroid carcinoma--who benefits? Langenbecks Arch Surg 1998;383:355-358 
  13. Roh JL, Park JY, Park CI. Total thyroidectomy plus neck dissection in differentiated papillary thyroid carcinoma patients: pattern of nodal metastasis, morbidity, recurrence, and postoperative levels of serum parathyroid hormone. Ann Surg 2007;245:604-610 
  14. Wada N, Duh QY, Sugino K, Iwasaki H, Kameyama K, Mimura T, et al. Lymph node metastasis from 259 papillary thyroid microcarcinomas: frequency, pattern of occurrence and recurrence, and optimal strategy for neck dissection. Ann Surg 2003;237:399-407 
  15. Cho BY, Choi HS, Park YJ, Lim JA, Ahn HY, Lee EK, et al. Changes in the clinicopathological characteristics and outcomes of thyroid cancer in Korea over the past four decades. Thyroid 2013;23:797-804 
  16. Mazzaferri EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer. Am J Med 1994;97:418-428 
  17. Su DH, Chang SH, Chang TC. The impact of locoregional recurrences and distant metastases on the survival of patients with papillary thyroid carcinoma. Clin Endocrinol (Oxf) 2015;82:286-294 
  18. Schlumberger MJ. Papillary and follicular thyroid carcinoma. N Engl J Med 1998;338:297-306 
  19. Liu W, Cheng R, Ma Y, Wang D, Su Y, Diao C, et al. Establishment and validation of the scoring system for preoperative prediction of central lymph node metastasis in papillary thyroid carcinoma. Sci Rep 2018;8:6962 
  20. Kwak JY, Kim EK, Kim MJ, Son EJ, Chung WY, Park CS, et al. Papillary microcarcinoma of the thyroid: predicting factors of lateral neck node metastasis. Ann Surg Oncol 2009;16:1348-1355 
  21. Lee SH, Lee SS, Jin SM, Kim JH, Rho YS. Predictive factors for central compartment lymph node metastasis in thyroid papillary microcarcinoma. Laryngoscope 2008;118:659-662 
  22. Suk HI, Lee SW, Shen D; Alzheimer's Disease Neuroimaging Initiative. Latent feature representation with stacked auto-encoder for AD/MCI diagnosis. Brain Struct Funct 2015;220:841-859 
  23. Zhang W, Li R, Deng H, Wang L, Lin W, Ji S, et al. Deep convolutional neural networks for multi-modality isointense infant brain image segmentation. Neuroimage 2015;108:214-224 
  24. Shin HC, Orton MR, Collins DJ, Doran SJ, Leach MO. Stacked autoencoders for unsupervised feature learning and multiple organ detection in a pilot study using 4D patient data. IEEE Trans Pattern Anal Mach Intell 2013;35:1930-1943 
  25. Choi WJ, Park JS, Kim KG, Kim SY, Koo HR, Lee YJ. Computerized analysis of calcification of thyroid nodules as visualized by ultrasonography. Eur J Radiol 2015;1;84:1949-1953 
  26. Li H, Weng J, Shi Y, Gu W, Mao Y, Wang Y, et al. An improved deep learning approach for detection of thyroid papillary cancer in ultrasound images. Sci Rep 2018;8:6600 
  27. Ma J, Wu F, Zhu J, Xu D, Kong D. A pre-trained convolutional neural network based method for thyroid nodule diagnosis. Ultrasonics 2017;73:221-230 
  28. Wu H, Deng Z, Zhang B, Liu Q, Chen J. Classifier model based on machine learning algorithms: application to differential diagnosis of suspicious thyroid nodules via sonography. AJR Am J Roentgenol 2016;207:859-864 
  29. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit 2016;12;770-778 
  30. Pan SJ, Yang Q. A survey on transfer learning. IEEE Trans Knowl Data Eng 2010;22:1345-1359 
  31. Moon KW. R statistics and graphs for medical papers. Seoul: Hannarae Publishing Co 2015 
  32. Team RC. A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing 2016 
  33. Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, et al. 2015 American Thyroid Association Management Guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association Guidelines Task Force on thyroid nodules and differentiated thyroid cancer. Thyroid 2016;26:1-133 
  34. Ito Y, Miyauchi A, Kihara M, Higashiyama T, Kobayashi K, Miya A. Patient age is significantly related to the progression of papillary microcarcinoma of the thyroid under observation. Thyroid 2014;24:27-34 
  35. Ito Y, Miyauchi A, Kobayashi K, Miya A. Prognosis and growth activity depend on patient age in clinical and subclinical papillary thyroid carcinoma. Endocr J 2014;61:205-213 
  36. Lundgren CI, Hall P, Dickman PW, Zedenius J. Clinically significant prognostic factors for differentiated thyroid carcinoma: a population-based, nested case-control study. Cancer 2006;106:524-531 
  37. Kim SY, Kwak JY, Kim EK, Yoon JH, Moon HJ. Association of preoperative US features and recurrence in patients with classic papillary thyroid carcinoma. Radiology 2015;277:574-583 
  38. Jeong EY, Kim HL, Ha EJ, Park SY, Cho YJ, Han M. Computer-aided diagnosis system for thyroid nodules on ultrasonography: diagnostic performance and reproducibility based on the experience level of operators. Eur Radiol 2019;29:1978-1985 
  39. Chi J, Walia E, Babyn P, Wang J, Groot G, Eramian M. Thyroid nodule classification in ultrasound images by fine-tuning deep convolutional neural network. J Digit Imaging 2017;30:477-486 
  40. Kim SY, Lee E, Nam SJ, Kim EK, Moon HJ, Yoon JH, et al. Ultrasound texture analysis: association with lymph node metastasis of papillary thyroid microcarcinoma. PLoS One 2017;12:e0176103