Leveraging network using controlled weight learning approach for thyroid cancer lymph node detection

• Autorzy: Kavitha Muthu Subash , Yudistira Novanto , Ahn Byeong-Cheol , Kurita Takio

Identyfikatory

DOI

10.1016/j.bbe.2021.10.003

• Języki publikacji: EN

Abstrakty

EN

Identification of cervical metastatic lymph nodes (LN) on I-131 post-ablation whole-body planar scans (WBS) for cancer staging is crucial for patients with papillary thyroid cancer (PTC). The existing deep network is plagued by instability in finding the under-represented LN classes on highly complex WBS, where end-to-end tuned models that can account for this uncertainty region for multi-class locations are needed. Hence, as a key contribution of this study, we designed a novel leveraging segmentation network with input guidance (LSIG) end-to-end training model without pre and post-processing features that can learn ideal parameter settings depending on the quantity of multiple-object instances. To improve the co-occurrence of classes and control the false positive regions, we proposed a re-weighting negative control (RNC) mechanism that combines two key components, namely the re-weighting (Rw) term and the negative control function (NcF). This unified approximation of weighted training would leverage the network to control and learn the desired weights of true positives towards the LN region. As an end-to-end network training, we utilize a Unet-like convolution neural network (CNN) model. The performance of the LSIG is compared with the CNN-based networks, based on the ground truth (GT) mask developed using post-ablation single-photon emission computed tomography (SPECT/CT). Furthermore, the effectiveness of the two components used in the LSIG framework is evaluated on WBS datasets. Our proposed LSIG with a fully guided (Fg) LS-FgCNN model yielded a superior performance with high AUC value of 94.9%, which is 14.6% higher than the previous network for PTC.

Słowa kluczowe

EN

uncertainty; multiclass; imbalance; co-occurrence; lymph node detection

PL

niepewność; nierównowaga; współwystępowanie; wykrywanie węzłów chłonnych

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2021
• Tom: Vol. 41, no. 4
• Strony: 1589--1600
• Opis fizyczny: Bibliogr. 48 poz., rys., tab., wykr.

Twórcy

autor

Kavitha Muthu Subash

• School of Information and Data Sciences, Nagasaki University, 1-14 Bunkyo-machi, Nagasaki, Japan

autor

Yudistira Novanto

• Intelligent System laboratory, Faculty of Computer Science, Brawijaya University, Indonesia

autor

Ahn Byeong-Cheol

• Department of Nuclear Medicine, School of Medicine, Kyungpook National University, Kyungpook National University Hospital, Daegu, South Korea

autor

Kurita Takio

• Graduate School of Advanced Science and Engineering, Hiroshima University, Higashi-hiroshima, Hiroshima, Japan

Bibliografia

• [1] Wang TS, Dubner S, Sznyter LA, Heller KS. Arch Otolaryngol 2004;130(1):110.
• [2] Pacini F, Schlumberger M, Dralle H, Elisei R, Smit J, Wiersinga WM. Eur J Endocrinol 2006;154(6):787–803.
• [3] OH, J.; Ahn, B.-C. Am J Nucl Med Mol Imaging. 2012, 2(6), 362–85.
• [4] Garger Y, Winfeld M, Friedman K, Blum MJ. Investig Med High Impact Case Rep 2016;4(1):1–7.
• [5] Hannoush ZC, Palacios JD, Kuker RA, Casula S. Case Rep in Endocrinol 2017:1–5.
• [6] Munn LL, Padera TP. Microvasc Res 2014;96:55–63.
• [7] Wong KK, Fig LM, Youssef E, Ferretti A, Rubello D, Gross MD. Endocr Rev 2014;35:717–46.
• [8] Kitasaka, T., Tsujimura, Y., Nakamura, Y., Mori, K., Suenaga, Y., Ito, M. et al., Automated Extraction of Lymph Nodes from 3-D Abdominal CT Images Using 3-D Minimum Directional Difference Filter. 2007.
• [9] Barbu A, Suehling M, Xu X, Liu D, Zhou SK, Comaniciu D. IEEE Trans Med Imaging 2012;31:240–50.
• [10] Wang H, Zhou Z, Li Y, Chen Z, Lu P, WangW, et al. EJNMMI Res 2017:7.
• [11] Jun W, Eo T, Kim T, Shin H, Hwang D, Bae SH, et al. Sci Rep 2018;8:1.
• [12] Commowick O, Gfegoire V, Malandain G. Oncol 2008;87:281–9.
• [13] Gorthi, S., Duay, N., V. and Houhou, Bach Cuadra, M., Schick, U., Becker, M.; et al,. IEEE J Sel Top Signal Process. 2009, 3, 135–147.
• [14] Dornheim, J., Seim, H., Preim, B., Hertel, I., Strauss, G. Segmentation of neck lymph nodes in CT datasets with sTable 3D mass-spring models. 2006.
• [15] Kavitha MS, Lee C-H, Shibudas K, Kurita T, Ahn B-C. Sci Rep 2020:10.
• [16] Zanjani, F.G., Zinger, S., With, P.H. Cancer detection in histopathology whole-slide images using conditional random fields on deep embedded spaces. 2018.
• [17] Bejnordi B, Veta M, Diest PJV, et al. JAMA 2017;318.
• [18] Wang, D.; Khosla, A.; Gargeya, R.; Irshad, H.; Beck, A.H. arXiv.org, 2016. [Online]. Available: https://arxiv.org/abs/1606.05718. 2020, 318.
• [19] Koh J, Lee E, Han K, et al. Sci Rep 2020:10.
• [20] Wang, L., Yang, S., S. Yang; et al., World J Surg Onc. 2019., 17.
• [21] Zhao, H., Liu, Q., JY. Lin, et al., Ann Transl Med. 2020., 8.
• [22] Yudistira N, Kavitha MS, Itabashi T, et al. Sci Rep 2020:10.
• [23] Wang Y, Guan Q, Lao I, Wang L, Wu Y, Li D, Ji Q, Wang Y, Zhu Y, Lu H, Xiang J. Ann Transl Med 2019;7:468.
• [24] Halicek, M., Shahedi, M., Little, J., V., et al., Sci Rep 2019, 9.
• [25] Diamant A, Chatterjee A, Vallières M, Shenouda G, Seuntjens J. Sci Rep 2019:9.
• [26] Halicek, M., Little, J., V. Wang, X., Patel, M., Griffith, C.C.; El-Deiry, M., Chen, A., Fei, B. Optical biopsy of head and neck cancer using hyperspectral imaging and convolutional neural networks, 2018.
• [27] ZhuW, Huang Y, Zeng L, Chen X, Liu Y, Qian Z, et al. Med Phys 2018;46:576–89.
• [28] Schlemper J, Oktay O, Schaap M, Heinrich M, Kainz B, Glocker B, Rueckert D. Med Image Anal 2019;53:197–207.
• [29] Gadosey, P.K., Li, Y., Agyekum, E., A., Zhang, T., Liu, Z., Yamak, P., Essaf, F. Diagnostics. 2020, 10.
• [30] Li H, Huang Y, Zhang Z. IEEE Access 2017;5:13665–76.
• [31] Ibragimov B, Xing L. Med Phys 2017;44:547–57.
• [32] Nikolov, S., Blackwell, S.; Mendes, R., Fauw, J., D. Meyer, C.; Hughes, C. et al., arXiv.org, 2018. [Online]. Available: https://arxiv.org/abs/1809.04430. [Accessed: 08-Sep-2020].
• [33] Ren, X., Xiang, L., H., Nie, D., Shao, Y., Zhang, H., Shen, D.; et al,. Med Phys 2018., 45, 2063–2075.
• [34] Johnson JM, Khoshgoftaar TMJ. Big Data 2019:6.
• [35] Huang, C., Li, Y., Loy, C.C., Tang, X. Learning deep representation for imbalanced classification.
• [36] Shen, L., Lin, Z., Huang, Q. Relay backpropagation for effective learning of deep convolutional neural networks.
• [37] Buda M, Maki A, Mazurowski M. Neural Netw 2018;106:249–59.
• [38] Zong W, Huang G, Chen Y. Neurocomputing 2013;101:229–42.
• [39] Ma, L., Lu, G., Wang, X., D. ans Qin, Chen, Z., G., Fei, B. Vis. Comput. Ind. Biomed. Art 2019, 2.
• [40] Elkan, C. The foundations of cost-sensitive learning. 2001.
• [41] Kahn H, Marshall AW. J Oper Res Soc 1953;1:263–78.
• [42] Lin, T.Y., Goyal, P., Girshick, R., He, K., Dollár, P. Focal loss for dense object detection. 2017.
• [43] Hen, B., C, Deng, W., Du, J. Noisy softmax: Improving the generalization ability of DCNN via postponing the early softmax saturation. 2017.
• [44] Dong, Q., Gong, S., Zhu, X. Class rectification hard mining for imbalanced deep learning. 2017.
• [45] Jo T, Japkowicz N. SIGKDD Explor. Newsl 2004;6:40–9.
• [46] Nguyen, T., Shih, M., Srivastava, D., Tirthapura, S., Xu, B. Stratified Random Sampling over Streaming and Stored Data. 2019.
• [47] Sudre, C., H. Li, W., Vercauteren, T.; Ourselin, S.; Cardoso, M. ArXiv abs/1707.03237 2017.
• [48] Crum, W.R., Camara, O., Hill, D., L., G., IEEE Trans on Med Imag 2006, 25, 451–1461.