{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T03:11:39Z","timestamp":1790651499445,"version":"4.1.0"},"posted":{"date-parts":[[2024,10,18]]},"group-title":"Bioengineering","reference-count":49,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2024,10,18]],"date-time":"2024-10-18T00:00:00Z","timestamp":1729209600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2024,10,18]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Computational models, particularly finite-element (FE) models, are essential for interpreting experimental data and predicting system behavior, especially when direct measurements are limited. A major challenge in tuning these models is the large number of parameters involved. Traditional methods, such as one-by-one sensitivity analyses, are time-consuming, subjective, and often return only a single set of parameter values, focusing on reproducing averaged data rather than capturing the full variability of experimental measurements.<\/jats:p>\n                <jats:p>In this study, we applied simulation-based inference (SBI) using neural posterior estimation (NPE) to tune an FE model of the human middle ear. The training dataset consisted of 10,000 FE simulations of stapes velocity, ear-canal (EC) input impedance, and absorbance, paired with seven FE parameter values randomly sampled within plausible ranges. The neural network learned the association between parameters and simulation outcomes, returning the probability distribution of parameter values that can reproduce experimental data.<\/jats:p>\n                <jats:p>Our approach successfully identified parameter sets that reproduced three experimental datasets simultaneously. By accounting for experimental noise and variability during training, the method provided a probability distribution of parameters, representing all valid combinations that could fit the data, rather than tuning to averaged values. The network demonstrated robustness to noise and exhibited an efficient learning curve due to the large training dataset.<\/jats:p>\n                <jats:p>SBI offers an objective alternative to laborious sensitivity analyses, providing probability distributions for each parameter and uncovering interactions between them. This method can be applied to any biological FE model, and we demonstrated its effectiveness using a middle-ear model. Importantly, it holds promise for objective differential diagnosis of conductive hearing loss by providing insight into the mechanical properties of the middle ear.<\/jats:p>","DOI":"10.1101\/2024.10.15.618553","type":"posted-content","created":{"date-parts":[[2024,10,18]],"date-time":"2024-10-18T07:25:36Z","timestamp":1729236336000},"source":"Crossref","is-referenced-by-count":2,"title":["From Simulations to Inference: Using Machine Learning to Tune Patient-Specific Finite-Element Models of the Middle Ear Towards Objective Diagnosis"],"prefix":"10.64898","author":[{"given":"Hamid","family":"Motallebzadeh","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael","family":"Deistler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Florian M.","family":"Sch\u00f6nleitner","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5154-8912","authenticated-orcid":false,"given":"Jakob H.","family":"Macke","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sunil","family":"Puria","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2024102207150913000_2024.10.15.618553v1.1","doi-asserted-by":"publisher","DOI":"10.1016\/j.heares.2012.05.001"},{"key":"2024102207150913000_2024.10.15.618553v1.2","doi-asserted-by":"publisher","DOI":"10.1007\/s10237-011-0355-6"},{"key":"2024102207150913000_2024.10.15.618553v1.3","doi-asserted-by":"publisher","DOI":"10.1111\/joa.13457"},{"key":"2024102207150913000_2024.10.15.618553v1.4","unstructured":"Cai, H. , Jackson, R. P. , Steel, C. R. , and Puria, S. (2012). \u201cComplex Twisting Motion at the Malleus-Incus Complex in a Finite-Element Model of Human Middle Ear,\u201d 6th Int. MEMRO Meet. (Taegu, South Korea)."},{"key":"2024102207150913000_2024.10.15.618553v1.5","doi-asserted-by":"publisher","DOI":"10.1007\/s10237-007-0094-x"},{"key":"2024102207150913000_2024.10.15.618553v1.6","doi-asserted-by":"publisher","DOI":"10.1016\/j.medengphy.2007.04.005"},{"key":"2024102207150913000_2024.10.15.618553v1.7","doi-asserted-by":"publisher","DOI":"10.1159\/000091815"},{"key":"2024102207150913000_2024.10.15.618553v1.8","unstructured":"CMSOL Blog, C. (2017). \u201cHow to Use Model Methods to Accelerate Your COMSOL\u00ae Workflow,\u201d COMSOL."},{"key":"2024102207150913000_2024.10.15.618553v1.9","unstructured":"COMSOL Blog, C. (2017). \u201cHow to Export Images Automatically After Solving Your Model,\u201d COMSOL."},{"key":"2024102207150913000_2024.10.15.618553v1.10","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1912789117"},{"key":"2024102207150913000_2024.10.15.618553v1.11","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmbbm.2008.05.008"},{"key":"2024102207150913000_2024.10.15.618553v1.12","doi-asserted-by":"publisher","DOI":"10.1016\/j.heares.2016.11.011"},{"key":"2024102207150913000_2024.10.15.618553v1.13","doi-asserted-by":"publisher","DOI":"10.1016\/0021-9290(80)90056-1"},{"key":"2024102207150913000_2024.10.15.618553v1.14","unstructured":"Dharmarajan, A. , Ravicz, M. E. , O\u2019Connor, K. N. , and Puria, S. (2020). \u201cMiddle-ear muscle forces could aid the diagnosis of otosclerosis with wideband tympanometry,\u201d (San Jose, CA). Presented at the Association for Research in Otolaryngology."},{"key":"2024102207150913000_2024.10.15.618553v1.15","unstructured":"Dobrev, I. , Kim, N. , Stenfelt, S. , Huber, A. M. , and R\u00f6\u00f6sli, C. (n.d.). \u201cExperimental and Simulation Evaluation of the Bone Conduction Surface Wave Propagation.,\u201d"},{"key":"2024102207150913000_2024.10.15.618553v1.16","unstructured":"Durkan, C. , Murray, I. , and Papamakarios, G. (2020). \u201cOn contrastive learning for likelihood-free inference,\u201d Int. Conf. Mach. Learn. (PMLR), pp. 2771\u20132781."},{"key":"2024102207150913000_2024.10.15.618553v1.17","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-34018-w"},{"key":"2024102207150913000_2024.10.15.618553v1.18","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.0603898104"},{"key":"2024102207150913000_2024.10.15.618553v1.19","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbiomech.2004.08.022"},{"key":"2024102207150913000_2024.10.15.618553v1.20","doi-asserted-by":"publisher","DOI":"10.1016\/j.heares.2018.06.019"},{"key":"2024102207150913000_2024.10.15.618553v1.21","unstructured":"Frei, W. (2014). \u201cUnderstanding Parallel Computing,\u201d COMSOL."},{"key":"2024102207150913000_2024.10.15.618553v1.22","doi-asserted-by":"publisher","DOI":"10.1097\/00129492-200205000-00008"},{"key":"2024102207150913000_2024.10.15.618553v1.23","doi-asserted-by":"publisher","DOI":"10.1016\/j.medengphy.2010.10.022"},{"key":"2024102207150913000_2024.10.15.618553v1.24","unstructured":"Greenberg, D. , Nonnenmacher, M. , and Macke, J. (2019). \u201cAutomatic posterior transformation for likelihood-free inference,\u201d Int. Conf. Mach. Learn. (PMLR), pp. 2404\u20132414."},{"key":"2024102207150913000_2024.10.15.618553v1.25","doi-asserted-by":"publisher","DOI":"10.1121\/1.4916200"},{"key":"2024102207150913000_2024.10.15.618553v1.26","doi-asserted-by":"publisher","DOI":"10.1121\/1.407347"},{"key":"2024102207150913000_2024.10.15.618553v1.27","doi-asserted-by":"publisher","DOI":"10.1121\/1.4935386"},{"key":"2024102207150913000_2024.10.15.618553v1.28","doi-asserted-by":"publisher","DOI":"10.1117\/1.JBO.20.5.051028"},{"key":"2024102207150913000_2024.10.15.618553v1.29","doi-asserted-by":"publisher","DOI":"10.1016\/j.bpj.2014.04.052"},{"key":"2024102207150913000_2024.10.15.618553v1.30","doi-asserted-by":"publisher","DOI":"10.1007\/s10162-015-0525-9"},{"key":"2024102207150913000_2024.10.15.618553v1.31","doi-asserted-by":"publisher","DOI":"10.1007\/s10162-015-0531-y"},{"key":"2024102207150913000_2024.10.15.618553v1.32","doi-asserted-by":"publisher","DOI":"10.1016\/j.heares.2016.07.018"},{"key":"2024102207150913000_2024.10.15.618553v1.33","doi-asserted-by":"publisher","DOI":"10.1121\/1.4828831"},{"key":"2024102207150913000_2024.10.15.618553v1.34","doi-asserted-by":"publisher","DOI":"10.1007\/s10162-017-0630-z"},{"key":"2024102207150913000_2024.10.15.618553v1.35","doi-asserted-by":"publisher","DOI":"10.1007\/s10162-016-0587-3"},{"key":"2024102207150913000_2024.10.15.618553v1.36","doi-asserted-by":"publisher","DOI":"10.1121\/10.0004218"},{"key":"2024102207150913000_2024.10.15.618553v1.37","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1720979115"},{"key":"2024102207150913000_2024.10.15.618553v1.38","doi-asserted-by":"publisher","DOI":"10.1142\/S0219519415500505"},{"key":"2024102207150913000_2024.10.15.618553v1.39","doi-asserted-by":"publisher","DOI":"10.1121\/10.0025768"},{"key":"2024102207150913000_2024.10.15.618553v1.40","unstructured":"O\u2019Connor, K. N. , Cai, H. , Gottlieb, P. , Steel, C. R. , and Puria, S. (2016). \u201cVarying the Material Properties of the Human Ossicular Chain in a 3D Finite-Element Model to Examine their Effects on Middle-Ear Sound Transmission,\u201d Int. Union Theor. Appl. Mech. -Symp. Adv. Biomech. Hear. (Stuttgart, Germany)."},{"key":"2024102207150913000_2024.10.15.618553v1.41","doi-asserted-by":"publisher","DOI":"10.1121\/1.5008741"},{"key":"2024102207150913000_2024.10.15.618553v1.42","unstructured":"Papamakarios, G. , and Murray, I. (2016). \u201cFast \u03b5-free inference of simulation models with bayesian conditional density estimation,\u201d Adv. Neural Inf. Process. Syst. Retrieved from https:\/\/proceedings.neurips.cc\/paper\/2016\/hash\/6aca97005c68f1206823815f66102863-Abstract.html."},{"key":"2024102207150913000_2024.10.15.618553v1.43","doi-asserted-by":"publisher","DOI":"10.1016\/j.heares.2019.02.011"},{"key":"2024102207150913000_2024.10.15.618553v1.44","doi-asserted-by":"publisher","DOI":"10.1115\/1.4052371"},{"key":"2024102207150913000_2024.10.15.618553v1.45","doi-asserted-by":"publisher","DOI":"10.1007\/s10162-010-0207-6"},{"key":"2024102207150913000_2024.10.15.618553v1.46","doi-asserted-by":"crossref","unstructured":"Tejero-Cantero, A. , Boelts, J. , Deistler, M. , Lueckmann, J.-M. , Durkan, C. , Gon\u00e7alves, P. J. , Greenberg, D. S. , et al. (2020, July 22). \u201cSBI -- A toolkit for simulation-based inference.,\u201d","DOI":"10.21105\/joss.02505"},{"key":"2024102207150913000_2024.10.15.618553v1.47","unstructured":"Winkler, S. T. , Sackmann, B. , Priwitzer, B. , and Lauxmann, M. (2024). \u201cExploring machine learning diagnostic decisions based on wideband immittance measurements for otosclerosis and disarticulation,\u201d J. Med. Artif. Intell. Retrieved from https:\/\/cdn.amegroups.cn\/journals\/jlpm\/files\/journals\/33\/articles\/9050\/public\/9050-PB1-7342-R1.pdf."},{"key":"2024102207150913000_2024.10.15.618553v1.48","doi-asserted-by":"publisher","DOI":"10.1504\/IJECB.2010.035260"},{"key":"2024102207150913000_2024.10.15.618553v1.49","doi-asserted-by":"publisher","DOI":"10.1007\/s10237-010-0268-9"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2024.10.15.618553","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T07:17:14Z","timestamp":1768461434000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.1101\/2024.10.15.618553"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,10,18]]},"references-count":49,"URL":"https:\/\/doi.org\/10.1101\/2024.10.15.618553","relation":{},"subject":[],"published":{"date-parts":[[2024,10,18]]},"subtype":"preprint"}}