복부 CT 영상에서 신장 로컬 가이드 맵을 활용한 평균-교사 모델 기반의 준지도학습을 통한 신장 종양 분할
Kidney Tumor Segmentation through Semi-supervised Learning Based on Mean Teacher Using Kidney Local Guided Map in Abdominal CT Images
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
부분신장절제술 전 수술 계획을 세우기 위해서는 신장 종양의 위치, 형태 및 수술 시 안전 마진 파악이 중요하므로 신장 종양을 정확히 분할하는 것이 필요하다. 그러나 신장 종양은 환자마다 위치 및 크기가 다양하며 소장과 비장 같은 주변 장기와 형태와 밝기값이 유사하여 신장 종양을 분할하는 것에 어려움이 있다. 본 논문에서는 레이블이 있는 데이터와 없는 데이터를 함께 사용하는 준지도학습 방법 중 하나인 평균-교사모델을 활용하여 신장의 여러 위치에서 발생하는 작은 크기의 신장 종양을 분할하기 위해 신장 위치 정보를 가지는 신장로컬 가이드 맵을 이용해 신장 종양에 집중하는 평균-교사 네트워크를 제안하고, 신장 종양의 크기에 따른 성능을 분석한다. 실험 결과, 제안 방법은 신장 주변에 존재하는 종양의 위치를 찾기 위해 신장 로컬 가이드 맵을 사용하여 신장의 국소 정보를 고려함으로써 75.24%의 F1-score를 보였다. 특히 분할이 어려운 작은 크기의종양에 대한 과소분할을 개선하였으며 nnU-Net보다 적은 양의 레이블 데이터를 사용하여도 13.9% 높은F1-score를 보였다.
Abstract
Accurate segmentation of the kidney tumor is necessary to identify shape, location and safety margin of tumor in abdominal CT images for surgical planning before renal partial nephrectomy. However, kidney tumor segmentation is challenging task due to the various sizes and locations of the tumor for each patient and signal intensity similarity to surrounding organs such as intestine and spleen. In this paper, we propose a semi-supervised learning-based mean teacher network that utilizes both labeled and unlabeled data using a kidney local guided map including kidney local information to segment small-sized kidney tumors occurring at various locations in the kidney, and analyze the performance according to the kidney tumor size. As a result of the study, the proposed method showed an F1-score of 75.24% by considering local information of the kidney using a kidney local guide map to locate the tumor existing around the kidney. In particular, under-segmentation of small-sized tumors which are difficult to segment was improved, and showed a 13.9%p higher F1-score even though it used a smaller amount of labeled data than nnU-Net.
참고문헌 (27)
- H. Sung, "Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries", A Cancer Journal for Clinicians, 71(3), pp.209~249 (2021)
- Z. Du, "Trends a nd projections of kidney cancer incidence at the global and national levels, 1990–2030 : A Bayesian age-period-cohort m odeling study", Biomarker Research, 16, pp.8~ (2020)
- S. Tangal, "Relat ion of neutrophil lymphocyte ratio with tumor characteristic s in localized kidney tumors", The New Journal of Urology, 13(1), pp.12~15 (2018)
- M. Sun, "Treatment management of small renal masses in the 21st century : A paradigm shift", Annals of S urgial. Oncology, 19, pp.2380~2387 (2012)
- G. Yang, "Automatic kidney segmentation in ct image s based on multi-atlas image registration", Proc. of the 36th Annual Int. Conf. of the IEEE Engineering in Medici ne and Biology Society, pp.5538~5541 (2014)
- D. K. Kim, "Two-year Analys is for Predicting Renal Function and Contralateral Hypertro phy after Robot-assisted Partial Nephrectomy : A Three-dime nsional Segmentation Technology Study", International Journal of Urology, 22, pp.1105~1111 (2015)
- 이재선, "복부 컴퓨터 단층촬영영상에서 다중 확률 아틀라스 기반 형상제한 그래프-컷을 사용한 신실질 자동 분할", 한국컴퓨터그래픽스학회논문지, 22(4), pp.11~19 (2016), (사)한국컴퓨터그래픽스학회[DOI]
- H. Wei, "Two-phase Framework for Automatic Kidney and Kidn ey Tumor Segmentation", 2019 Kidney Tumor Segmentatio n Challenge: KiTS19 (2019)
- X. Xie, "SERU : A c ascaded SE-ResNeXT U-Net for kidney and tumor segment ation", Concurrency and Computation: Practice and Experience, 32(14) (2020)
- J. Wen, "Squee ze-and-Excitation Encoder-Decoder Network for Kidney and Kidney Tumor Segmentation in CT images", Kidney and K idney Tumor Segmentation. KiTS 2021. Lecture Notes in C omputer Science, 13168 (2022), Springer, Cham.
- F. Isensee, "An attempt at beating the 3D U-Net", ArXiv abs, 2019 (1908)
- Xiaoshuai Hou, "Cascaded Semantic Segmentation for Kidney and Tumor", Submissions to the 2019 Kidney Tumor Segmentation Challenge: KiTS19 (2019), University of Minnesota Libraries Publishing[DOI]
- Jun Ma, "Solution to the Kidney Tumor Segmentation Challenge 2019", Submissions to the 2019 Kidney Tumor Segmentation Challenge: KiTS19 (2019), University of Minnesota Libraries Publishing[DOI]
- M. J. Willemink, "Preparing medical imaging data for machine learning", Radiology, 295 (2020)
- S. Y. Huang, "Fully Convolutional Network for the Semantic Segmentation of Medical Images : A Survey", Diagnostics, 12 (2022)
- Z. Yang, "A Densely Co nnected Network Based on U-Net for Medical Image Segm entation", ACM Trans Multimed Comput Commun Appl, 17 (2021)
- L. Liu, "Anatomy-aided deep learning for medical image segmentati on : A review", Phys Med Biol, 66 (2021)
- D. Karimi, "Convolut ion-Free Medical Image Segmentation Using Transformers", Lect Notes Comput Sci(Including Subser Lect Notes Artif Intell Lect Notes Bioinformatics), 12901 (2021)
- Y. Wang, "Double-Uncertainty Weighted Method for S emi-supervised Learning", International Conference on Medi cal Image Computing and Computer-Assisted Intervention (2020)
- J. Wang, "Rethinking Bayesian Deep Learning Methods for Semi-Supervised Volumetric Medical Image Segmentation", 2022 IEEE, 2022, 182~190
- Y. Tang, "Cons istency and adversarial semi-supervised learning for medical image segmentation", Computers in Biology and Medicine, 161 (2023)
- L. Hu, "Semi-supervised NPC segmentation with uncertainty and attention guided consistency", Knowled ge-Based Systems, 239 (2022)
- A. Tarvainen, "Mean teachers are better r ole models: Weight-averaged consistency targets improve se mi-supervised deep learning results", Adv Neural Inf Proces s Syst (2017)
- O. Ronneberger, "U-net : Convolu tional networks for biomedical image segmentation", Interna tional Conference on Medical Image Computing and Comp uter-Assisted Intervention–MICCAI 2015, 9351, pp.234~241 (2015)
- F. Isensee, "nnU-Net : Self-adapting framew ork for u-net-based medical image segmentation", arXiv pre print arXiv : 1809. 10486, ~ (2018)
- E. Yang, "3D multi-scale residual fully convolutional neural network for s egmentation of extremely large-sized kidney tumor", Compu ter Methods and Programs in Biomedicine, 215 (2022)
- R. R. Selvaraju, "Grad-CAM : Visual Explanations fro m Deep Networks via Gradient-Based Localization", 2017 I EEE International Conference on Computer Vision (ICCV), pp.618~626 (2017)