Abstract
In clinical medicine, physicians often rely on information derived from medical imaging systems, such as image data for diagnosis. To detect disease early, physicians extract essential information from data manually to distinguish accurately between positive and negative cases of disease. In recent years, deep learning (DL) has been used for this purpose, attracting the attention of prominent researchers because of its excellent performance. Consequently, DL and other artificial intelligence (AI) technologies are expected to develop further through integration with statistical and other approaches. Here, we examine biliary atresia (BA), a rare disease that affects primarily infants. Our study focuses on the identification of BA from image data (stool images of BA patients). Using AI and statistical approaches, we propose a machine learning classifier (model) for accurate diagnosis, efficient classification, and early detection of BA after exposure to limited training data. In an initial study, we used the subspace pattern recognition method for the development of a similar classifier. In this study, we propose the development of a filter based on the subspace method and a statistical approach. The filter enables the classifier to extract essential information from image data and discriminate efficiently between BA and non-BA patients.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Chen SM, Chang MH, Du JC, Lin CC, Chen AC, Lee HC, Lau BH, Yang YJ, Wu TC, Chu CH, Lai MW, Chen HL, The Taiwan Infant Stool Color Card Study Group (2006) Screening for biliary atresia by infant stool color card in Taiwan. Pediatrics 117(4):1147–1154
Choi BC (1998) Slopes of a receiver operating characteristic curve and likelihood ratios for a diagnostic test. Am J Epidemiol 148(11):1127–1132
DeLong ER, DeLong DM, Clarke-Pearson DL (1988) Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44(3):837–845
Diaz-Chito K, Ferri FJ, Hernández-Sabaté A (2018) An overview of incremental feature extraction methods based on linear subspaces. Knowl Based Syst 145:219–235
Efron B (1979) Bootstrap methods: another look at the jackknife. Ann Stat 7(1):1–26
Efron B (1981) Nonparametric estimates of standard error: the jackknife, the bootstrap and other methods. Biometrika 68(3):589–599
Franciscovich A, Vaidya D, Doyle J, Bolinger J, Capdevila M, Rice M, Hancock L, Mahr T, Mogul DB (2015) PoopMD, a mobile health application, accurately identifies infant acholic stools. PLoS One 10(7):e0132270
Goodfellow I, Bengio Y, Courville A (2016) Deep learning. The MIT Press, Cambridge, Massachusetts
Gu YH, Yokoyama K, Mizuta K, Tsuchioka T, Kudo T, Sasaki H, Nio M, Tang J, Ohkubo T, Matsui A (2015) Stool color card screening for early detection of biliary atresia and long-term native liver survival: a 19-year cohort study in Japan. J Pediatr 166(4):897–902.e1
Hartley JL, Davenport M, Kelly DA (2009) Biliary atresia. Lancet 374(9702):1704–1713
Hastie T, Tibshirani R, Friedman J (2001) The elements of statistical learning, vol 1. Springer, Berlin
Hoshino E, Hayashi K, Suzuki M, Obatake M, Urayama KY, Nakano S, Taura Y, Nio M, Takahashi O (2017) An iPhone application using a novel stool color detection algorithm for biliary atresia screening. Pediatr Surg Int 33(10):1115–1121
Hsiao CH, Chang MH, Chen HL, Lee HC, Wu TC, Lin CC, Yang YJ, Chen AC, Tiao MM, Lau BH, Chu CH, Lai MW, the Taiwan Infant Stool Color Card Study Group (2008) Universal screening for biliary atresia using an infant stool color card in Taiwan. Hepatology 47(4):1233–1240
Ihaka R, Gentleman R (1996) R: a language for data analysis and graphics. J Comput Graph Stat 5(3):299–314
Jimenez-Rivera C, Jolin-Dahel KS, Fortinsky KJ, Gozdyra P, Benchimol EI (2013) International incidence and outcomes of biliary atresia. J Pediatr Gastroenterol Nutr 56(4):344–354
Kasai M, Kimura S, Asakura Y, Suzuki H, Taira Y, Ohashi E (1968) Surgical treatment of biliary atresia. J Pediatr Surg 3(6):665–675
Li Y, Lam F, Clifford B, Liang ZP (2017) A subspace approach to spectral quantification for MR spectroscopic imaging. IEEE Trans Biomed Eng 64(10):2486–2489
Lu L, Zheng Y, Carneiro G, Yang L (eds) (2017) Deep learning and convolutional neural networks for medical image computing. Advances in Computer Vision and Pattern Recognition. Springer, Switzerland
Matsui A, Dodoriki M (1995) Screening for biliary atresia. Lancet 345(8958):1181
Oja E (1983) Subspace methods of pattern recognition. Research Studies Press, England
R Core Team (2018) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/
Schreiber RA, Masucci L, Kaczorowski J, Collet JP, Lutley P, Espinosa V, Bryan S (2014) Home-based screening for biliary stresia using infant stool colour cards: a large-scale prospective cohort study and cost-effectiveness analysis. J Med Screen 21(3):126–132
Urbanek S (2014) jpeg: Read and write JPEG images. R package version 0.1–8. https://CRAN.R-project.org/package=jpeg
Watanabe S, Pakvasa N (1973) Subspace method of pattern recognition. In: Proceedings of the 1st International Joint Conference on Pattern Recognition, pp 25–32
Watanabe S, Lambert PF, Kulikowski CA, Buxton JL, Walker R (1967) Evaluation and selection of variables in pattern recognition. In: Tou J (ed) Computer and Information Science II. Academic Press, New York, pp 91–122
Wu F, Jing XY, Wu S, Gao G, Ge Q, Wang R (2018) “Like charges repulsion and opposite charges attraction” law based multilinear subspace analysis for face recognition. Knowl Based Syst 149:76–87
Xie L, Yin M, Wang L, Tan F, Yin G (2018) Matrix regression preserving projections for robust feature extraction. Knowl Based Syst 161:35–46
Acknowledgements
We wish to thank Shinsuke Ito, Takashi Taguchi, Dr. Yusuke Yamane, Ms. Saeko Hishinuma, and Dr. Saeko Hirai for their professional advice. In addition, we wish to acknowledge the biliary atresia patients’ community (BA no kodomowo mamorukai) for their generous support toward this project.
Funding
The funding organization does not have any role in the design and conduct of this study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
All the authors in this study declare that they do not have anything to disclose regarding conflict of interest.
Ethics statement
This study involved the use of de-identified photographs and was also approved by the Institutional Review Board.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work was supported by Japan Society for the Promotion of Science KAKENHI Grant Number 18H03336.
Rights and permissions
About this article
Cite this article
Hayashi, K., Hoshino, E., Suzuki, M. et al. Early identification of biliary atresia using subspace and the bootstrap methods. Adv Data Anal Classif 17, 163–179 (2023). https://doi.org/10.1007/s11634-022-00493-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11634-022-00493-8