{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,17]],"date-time":"2026-07-17T18:07:30Z","timestamp":1784311650954,"version":"3.55.0"},"reference-count":36,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2022,7,29]],"date-time":"2022-07-29T00:00:00Z","timestamp":1659052800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,7,29]],"date-time":"2022-07-29T00:00:00Z","timestamp":1659052800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100001663","name":"Volkswagen Foundation","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001663","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001652","name":"Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100001652","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Bioinformatics"],"published-print":{"date-parts":[[2022,12]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Background<\/jats:title><jats:p>Reference intervals represent the expected range of physiological test results in a healthy population and are essential to support medical decision making. Particularly in the context of pediatric reference intervals, where recruitment regulations make prospective studies challenging to conduct, indirect estimation strategies are becoming increasingly important. Established indirect methods enable robust identification of the distribution of \u201chealthy\u201d samples from laboratory databases, which include unlabeled pathologic cases, but are currently severely limited when adjusting for essential patient characteristics such as age. Here, we propose the use of mixture density networks (MDN) to overcome this problem and model all parameters of the mixture distribution in a single step.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Estimated reference intervals from varying settings with simulated data demonstrate the ability to accurately estimate latent distributions from unlabeled data using different implementations of MDNs. Comparing the performance with alternative estimation approaches further highlights the importance of modeling the mixture component weights as a function of the input in order to avoid biased estimates for all other parameters and the resulting reference intervals. We also provide a strategy to generate partially customized starting weights to improve proper identification of the latent components. Finally, the application on real-world hemoglobin samples provides results in line with current gold standard approaches, but also suggests further investigations with respect to adequate regularization strategies in order to prevent overfitting the data.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>Mixture density networks provide a promising approach capable of extracting the distribution of healthy samples from unlabeled laboratory databases while simultaneously and explicitly estimating all parameters and component weights as non-linear functions of the covariate(s), thereby allowing the estimation of age-dependent reference intervals in a single step. Further studies on model regularization and asymmetric component distributions are warranted to consolidate our findings and expand the scope of applications.<\/jats:p><\/jats:sec>","DOI":"10.1186\/s12859-022-04846-0","type":"journal-article","created":{"date-parts":[[2022,7,29]],"date-time":"2022-07-29T09:02:47Z","timestamp":1659085367000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":9,"title":["Mixture density networks for the indirect estimation of reference intervals"],"prefix":"10.1186","volume":"23","author":[{"given":"Tobias","family":"Hepp","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jakob","family":"Zierk","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Manfred","family":"Rauh","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Markus","family":"Metzler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sarem","family":"Seitz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2022,7,29]]},"reference":[{"key":"4846_CR1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0149856","volume":"11","author":"U Rohr","year":"2016","unstructured":"Rohr U, Binder C, Dieterle T, Giusti F, Messina C, Toerien E, et al. The value of in vitro diagnostic testing in medical practice: a status report. PLoS ONE. 2016;11: e0149856.","journal-title":"PLoS ONE"},{"key":"4846_CR2","doi-asserted-by":"crossref","unstructured":"Solberg H. Approved recommendation (1987) on the theory of reference values. Part 5. Statistical treatment of collected reference values. Clin Chim Acta. 1987;170(2-3):S13\u2013S32.","DOI":"10.1016\/0009-8981(87)90151-3"},{"key":"4846_CR3","unstructured":"CLSI, IFCC. C28-A3: Defining, establishing and verifying reference intervals in the clinical laboratory: approved guideline\u2014third edition; 2008."},{"issue":"12","key":"4846_CR4","doi-asserted-by":"publisher","first-page":"1893","DOI":"10.1515\/cclm-2016-0793","volume":"54","author":"J Henny","year":"2016","unstructured":"Henny J, Vassault A, Boursier G, Vukasovic I, Brguljan PM, Lohmander M, et al. Recommendation for the review of biological reference intervals in medical laboratories. Clin Chem Lab Med. 2016;54(12):1893\u2013900.","journal-title":"Clin Chem Lab Med"},{"issue":"8","key":"4846_CR5","doi-asserted-by":"publisher","first-page":"1033","DOI":"10.1515\/CCLM.2007.249","volume":"45","author":"R Haeckel","year":"2007","unstructured":"Haeckel R, Wosniok W, Arzideh F. A plea for intra-laboratory reference limits. Part 1. General considerations and concepts for determination. Clin Chem Lab Med. 2007;45(8):1033\u201342.","journal-title":"Clin Chem Lab Med"},{"issue":"1\u20132","key":"4846_CR6","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1016\/S0009-8981(03)00133-5","volume":"334","author":"PS Horn","year":"2003","unstructured":"Horn PS, Pesce AJ. Reference intervals: an update. Clin Chim Acta. 2003;334(1\u20132):5\u201323.","journal-title":"Clin Chim Acta"},{"issue":"5","key":"4846_CR7","doi-asserted-by":"publisher","first-page":"808","DOI":"10.1373\/clinchem.2012.183483","volume":"58","author":"F Ceriotti","year":"2012","unstructured":"Ceriotti F. Establishing pediatric reference intervals: a challenging task. Clin Chem. 2012;58(5):808\u201310.","journal-title":"Clin Chem"},{"issue":"4","key":"4846_CR8","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1258\/0004563041201572","volume":"41","author":"RS Brinkworth","year":"2004","unstructured":"Brinkworth RS, Whitham E, Nazeran H. Establishment of paediatric biochemical reference intervals. Ann Clin Biochem. 2004;41(4):321\u20139.","journal-title":"Ann Clin Biochem"},{"issue":"1","key":"4846_CR9","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1515\/cclm-2018-0073","volume":"57","author":"GR Jones","year":"2019","unstructured":"Jones GR, Haeckel R, Loh TP, Sikaris K, Streichert T, Katayev A, et al. Indirect methods for reference interval determination-review and recommendations. Clin Chem Lab Med (CCLM). 2019;57(1):20\u20139.","journal-title":"Clin Chem Lab Med (CCLM)"},{"issue":"4","key":"4846_CR10","doi-asserted-by":"publisher","first-page":"659","DOI":"10.1515\/CCLM.2011.114","volume":"49","author":"F Arzideh","year":"2011","unstructured":"Arzideh F, Wosniok W, Haeckel R. Indirect reference intervals of plasma and serum thyrotropin (TSH) concentrations from intra-laboratory data bases from several German and Italian medical centres. Clin Chem Lab Med. 2011;49(4):659\u201364.","journal-title":"Clin Chem Lab Med"},{"key":"4846_CR11","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41598-020-58749-2","volume":"10","author":"J Zierk","year":"2020","unstructured":"Zierk J, Arzideh F, Kapsner L, Prokosch HU, Metzler M, Rauh M. Reference interval estimation from mixed distributions using truncation points and the Kolmogorov\u2013Smirnov distance (kosmic). Sci Rep. 2020;10:1\u20138.","journal-title":"Sci Rep"},{"key":"4846_CR12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41598-021-95301-2","volume":"11","author":"T Ammer","year":"2021","unstructured":"Ammer T, Sch\u00fctzenmeister A, Prokosch HU, Rauh M, Rank CM, Zierk J. refineR: a novel algorithm for reference interval estimation from real-world data. Sci Rep. 2021;11:1\u201317.","journal-title":"Sci Rep"},{"issue":"7","key":"4846_CR13","doi-asserted-by":"publisher","first-page":"964","DOI":"10.1373\/clinchem.2015.239731","volume":"61","author":"J Zierk","year":"2015","unstructured":"Zierk J, Arzideh F, Rechenauer T, Haeckel R, Rascher W, Metzler M, et al. Age- and sex-specific dynamics in 22 hematologic and biochemical analytes from birth to adolescence. Clin Chem. 2015;61(7):964\u201373.","journal-title":"Clin Chem"},{"issue":"1","key":"4846_CR14","doi-asserted-by":"publisher","first-page":"102","DOI":"10.1515\/cclm-2016-0318","volume":"55","author":"J Zierk","year":"2017","unstructured":"Zierk J, Arzideh F, Haeckel R, Cario H, Fr\u00fchwald MC, Gro\u00df HJ, et al. Pediatric reference intervals for alkaline phosphatase. Clin Chem Lab Med (CCLM). 2017;55(1):102\u201310.","journal-title":"Clin Chem Lab Med (CCLM)"},{"key":"4846_CR15","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12859-020-03853-3","volume":"21","author":"T Hepp","year":"2020","unstructured":"Hepp T, Zierk J, Rauh M, Metzler M, Mayr A. Latent class distributional regression for the estimation of non-linear reference limits from contaminated data sources. BMC Bioinform. 2020;21:1\u201315.","journal-title":"BMC Bioinform"},{"issue":"3","key":"4846_CR16","doi-asserted-by":"publisher","first-page":"507","DOI":"10.1111\/j.1467-9876.2005.00510.x","volume":"54","author":"RA Rigby","year":"2005","unstructured":"Rigby RA, Stasinopoulos DM. Generalized additive models for location, scale and shape. J Roy Stat Soc Ser C (Appl Stat). 2005;54(3):507\u201354.","journal-title":"J Roy Stat Soc Ser C (Appl Stat)"},{"issue":"1","key":"4846_CR17","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1111\/j.2517-6161.1977.tb01600.x","volume":"39","author":"AP Dempster","year":"1977","unstructured":"Dempster AP, Laird NM, Rubin DB. Maximum likelihood from incomplete data via the EM algorithm. J Roy Stat Soc Ser B (Methodol). 1977;39(1):1\u201322.","journal-title":"J Roy Stat Soc Ser B (Methodol)"},{"key":"4846_CR18","unstructured":"Bishop CM. Mixture density networks;. Tech. Report at https:\/\/research.aston.ac.uk\/en\/publications\/mixture-density-networks (1994)."},{"issue":"4","key":"4846_CR19","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1007\/BF02478259","volume":"5","author":"WS McCulloch","year":"1943","unstructured":"McCulloch WS, Pitts W. A logical calculus of the ideas immanent in nervous activity. Bull Math Biophys. 1943;5(4):115\u201333.","journal-title":"Bull Math Biophys"},{"issue":"6","key":"4846_CR20","doi-asserted-by":"publisher","first-page":"386","DOI":"10.1037\/h0042519","volume":"65","author":"F Rosenblatt","year":"1958","unstructured":"Rosenblatt F. The perceptron: a probabilistic model for information storage and organization in the brain. Psychol Rev. 1958;65(6):386.","journal-title":"Psychol Rev"},{"key":"4846_CR21","unstructured":"PEDREF reference interval initiative;. https:\/\/www.pedref.org\/index.html. Accessed: 2021-12-01."},{"issue":"1","key":"4846_CR22","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1080\/00401706.1967.10490438","volume":"9","author":"AC Cohen","year":"1967","unstructured":"Cohen AC. Estimation in mixtures of two normal distributions. Technometrics. 1967;9(1):15\u201328.","journal-title":"Technometrics"},{"key":"4846_CR23","doi-asserted-by":"publisher","DOI":"10.1007\/978-94-009-5897-5","volume-title":"Finite mixture distributions. Monographs on applied probability and statistics","author":"B Everitt","year":"1981","unstructured":"Everitt B, Hand D. Finite mixture distributions. Monographs on applied probability and statistics. London: Chapman and Hall; 1981."},{"key":"4846_CR24","unstructured":"McLachlan G, Peel D. Finite mixture models. Wiley Series in Probability and Statistics. Wiley; 2004."},{"issue":"2","key":"4846_CR25","doi-asserted-by":"publisher","first-page":"249","DOI":"10.1007\/BF01897167","volume":"5","author":"WS DeSarbo","year":"1988","unstructured":"DeSarbo WS, Cron WL. A maximum likelihood methodology for clusterwise linear regression. J Classif. 1988;5(2):249\u201382.","journal-title":"J Classif"},{"issue":"1","key":"4846_CR26","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1007\/BF01202266","volume":"12","author":"M Wedel","year":"1995","unstructured":"Wedel M, DeSarbo WS. A mixture likelihood approach for generalized linear models. J Classif. 1995;12(1):21\u201355.","journal-title":"J Classif"},{"issue":"8","key":"4846_CR27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.18637\/jss.v011.i08","volume":"11","author":"F Leisch","year":"2004","unstructured":"Leisch F. FlexMix: a general framework for finite mixture models and latent class regression in R. J Stat Softw. 2004;11(8):1\u201318.","journal-title":"J Stat Softw"},{"key":"4846_CR28","volume-title":"Pattern recognition and machine learning","author":"CM Bishop","year":"2006","unstructured":"Bishop CM. Pattern recognition and machine learning. New York: Springer; 2006."},{"key":"4846_CR29","unstructured":"R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria; 2021. Available from: https:\/\/www.R-project.org\/."},{"key":"4846_CR30","volume-title":"Numerical optimization","author":"J Nocedal","year":"2006","unstructured":"Nocedal J, Wright SJ. Numerical optimization. 2nd ed. New York, NY, USA: Springer; 2006.","edition":"2"},{"key":"4846_CR31","unstructured":"Nash JC. Compact numerical methods for computers: linear algebra and function minimisation. Taylor & Francis; 1990. Available from: https:\/\/books.google.ne\/books?id=M9hTn3UAheQC."},{"key":"4846_CR32","unstructured":"Allaire J, Tang Y. tensorflow: R Interface to \u2019TensorFlow\u2019; 2021. R package version 2.5.0. Available from: https:\/\/CRAN.R-project.org\/package=tensorflow."},{"key":"4846_CR33","unstructured":"Keydana S. tfprobability: Interface to \u2019TensorFlow Probability\u2019; 2021. R package version 0.12.0.0. Available from: https:\/\/CRAN.R-project.org\/package=tfprobability."},{"key":"4846_CR34","unstructured":"Kingma DP, Ba J. Adam: A Method for Stochastic Optimization;. Preprint at https:\/\/arxiv.org\/abs\/1412.6980 [v9] (2017)."},{"key":"4846_CR35","unstructured":"Glorot X, Bengio Y. Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the thirteenth international conference on artificial intelligence and statistics. JMLR Workshop and Conference Proceedings; 2010. p. 249\u2013256."},{"issue":"7","key":"4846_CR36","doi-asserted-by":"publisher","first-page":"1","DOI":"10.18637\/jss.v023.i07","volume":"23","author":"DM Stasinopoulos","year":"2007","unstructured":"Stasinopoulos DM, Rigby RA, et al. Generalized additive models for location scale and shape (GAMLSS) in R. J Stat Softw. 2007;23(7):1\u201346.","journal-title":"J Stat Softw"}],"container-title":["BMC Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-022-04846-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12859-022-04846-0\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-022-04846-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,30]],"date-time":"2024-09-30T04:57:57Z","timestamp":1727672277000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcbioinformatics.biomedcentral.com\/articles\/10.1186\/s12859-022-04846-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,7,29]]},"references-count":36,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2022,12]]}},"alternative-id":["4846"],"URL":"https:\/\/doi.org\/10.1186\/s12859-022-04846-0","relation":{},"ISSN":["1471-2105"],"issn-type":[{"value":"1471-2105","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,7,29]]},"assertion":[{"value":"14 January 2022","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"15 July 2022","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 July 2022","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The study was approved by the institutional review board of the University Erlangen-Nuremberg (reference number 216_21 Bc).","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare that they have no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"307"}}