• Simulation-based inference for...

Treffer: Simulation-based inference for subject-specific tuning of middle ear finite-element models towards personalized objective diagnosis.

Title:
Simulation-based inference for subject-specific tuning of middle ear finite-element models towards personalized objective diagnosis.
Authors:
Motallebzadeh H; Department of Communication Sciences and Disorders, California State University, Sacramento, CA, USA. h.motallebzadeh@csus.edu.; Department of BioMedical Engineering, McGill University, Montreal, QC, Canada. h.motallebzadeh@csus.edu., Deistler M; Machine Learning in Science, University of Tübingen, Tübingen, Germany.; Tübingen AI Center, Tübingen, Germany., Schönleitner FM; TUM School of Computation, Information and Technology, Garching, Germany., Macke JH; Machine Learning in Science, University of Tübingen, Tübingen, Germany.; Tübingen AI Center, Tübingen, Germany.; Department Empirical Inference, Max Planck Institute for Intelligent Systems, Tübingen, Germany., Puria S; Eaton-Peabody Laboratories, Massachusetts Eye and Ear, Boston, USA.; Speech and Hearing Bioscience and Technology, Harvard University Graduate School of Arts and Sciences, Cambridge, USA.
Source:
Scientific reports [Sci Rep] 2025 Nov 03; Vol. 15 (1), pp. 38364. Date of Electronic Publication: 2025 Nov 03.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Nature Publishing Group Country of Publication: England NLM ID: 101563288 Publication Model: Electronic Cited Medium: Internet ISSN: 2045-2322 (Electronic) Linking ISSN: 20452322 NLM ISO Abbreviation: Sci Rep Subsets: MEDLINE
Imprint Name(s):
Original Publication: London : Nature Publishing Group, copyright 2011-
MeSH Terms:
Ear, Middle*/physiology , Finite Element Analysis* , Computer Simulation* , Precision Medicine*/methods, Humans ; Neural Networks, Computer
References:
J Acoust Soc Am. 2024 Apr 1;155(4):2769-2785. (PMID: 38662609)
Med Eng Phys. 2017 Feb;40:65-74. (PMID: 27989383)
J Acoust Soc Am. 2016 Jan;139(1):104-17. (PMID: 26827009)
Biophys J. 2014 Jul 1;107(1):233-41. (PMID: 24988357)
J Anat. 2021 Oct;239(4):771-781. (PMID: 34057736)
Hear Res. 2019 Jul;378:126-138. (PMID: 30878271)
Biomech Model Mechanobiol. 2011 Oct;10(5):713-26. (PMID: 21061141)
Hear Res. 2007 Aug;230(1-2):24-33. (PMID: 17517484)
Biomech Model Mechanobiol. 2013 Aug;12(4):705-15. (PMID: 22965177)
J Biomech. 2005 Sep;38(9):1807-15. (PMID: 16023467)
Sci Rep. 2023 May 5;13(1):7329. (PMID: 37147426)
J Assoc Res Otolaryngol. 2017 Feb;18(1):25-48. (PMID: 27718037)
J Acoust Soc Am. 2013 Dec;134(6):4427. (PMID: 25669254)
J Acoust Soc Am. 1993 Nov;94(5):2617-38. (PMID: 8270739)
J Acoust Soc Am. 2021 Apr;149(4):2711. (PMID: 33940924)
J Acoust Soc Am. 2018 Jun;143(6):3418. (PMID: 29960477)
Proc Natl Acad Sci U S A. 2018 May 29;115(22):5762-5767. (PMID: 29760098)
J Mech Behav Biomed Mater. 2009 Jan;2(1):82-92. (PMID: 19627811)
J Assoc Res Otolaryngol. 2015 Aug;16(4):433-46. (PMID: 26040214)
Hear Res. 2001 Feb;152(1-2):100-9. (PMID: 11223285)
Med Eng Phys. 2011 Apr;33(3):330-9. (PMID: 21112232)
J Biomech. 1980;13(7):559-64. (PMID: 7400184)
Biomech Model Mechanobiol. 2012 Jul;11(6):829-40. (PMID: 22038402)
Biomech Model Mechanobiol. 2008 Oct;7(5):387-94. (PMID: 17710457)
J Assoc Res Otolaryngol. 2015 Oct;16(5):547-67. (PMID: 26197870)
J Acoust Soc Am. 2015 Apr;137(4):1877-87. (PMID: 25920840)
Med Eng Phys. 2018 Apr;54:14-21. (PMID: 29477280)
Med Eng Phys. 2008 Apr;30(3):358-66. (PMID: 17553724)
J Biomech Eng. 2022 Mar 1;144(3):. (PMID: 34505125)
PLoS One. 2024 Apr 10;19(4):e0298535. (PMID: 38598472)
J Acoust Soc Am. 2022 Nov;152(5):2769. (PMID: 36456266)
Proc Natl Acad Sci U S A. 2020 Dec 1;117(48):30055-30062. (PMID: 32471948)
J Biomech Eng. 2014 Aug;136(8):. (PMID: 24828880)
J Assoc Res Otolaryngol. 2017 Oct;18(5):671-686. (PMID: 28721606)
J Biomech Eng. 2007 Dec;129(6):913-18. (PMID: 18067396)
Hear Res. 2016 Nov;341:19-30. (PMID: 27496538)
Proc Natl Acad Sci U S A. 2006 Dec 26;103(52):19743-8. (PMID: 17170142)
J Biomed Opt. 2015 May;20(5):051028. (PMID: 25652791)
Hear Res. 2012 Aug;290(1-2):45-54. (PMID: 22583920)
Audiol Neurootol. 2006;11(3):183-97. (PMID: 16514236)
Hear Res. 1996 Aug;97(1-2):30-45. (PMID: 8844184)
J Acoust Soc Am. 2017 Nov;142(5):2836. (PMID: 29195482)
Hear Res. 2018 Jul;364:96-103. (PMID: 29602590)
Grant Information:
R21 DC020274 United States DC NIDCD NIH HHS; PJT-189955 Canada CAPMC CIHR; R21DC020274 NIH/NIDCD
Contributed Indexing:
Keywords: Finite-element method; Patient-specific models; Simulation-based inference; Tuning mechanistic models
Entry Date(s):
Date Created: 20251103 Date Completed: 20251103 Latest Revision: 20251107
Update Code:
20251107
PubMed Central ID:
PMC12583476
DOI:
10.1038/s41598-025-22164-2
PMID:
41184342
Database:
MEDLINE

Weitere Informationen

Computational models, particularly finite-element (FE) models, are essential for interpreting experimental data and predicting system behavior, especially when direct measurements are limited. Tuning these models is particularly challenging when a large number of parameters are involved. Traditional methods, such as sensitivity analyses, are time-consuming and often provide only a single set of parameter values, focusing on reproducing averaged trends rather than capturing experimental variability. New approaches are needed to make computational models more adaptable to patient-specific clinical applications. We applied simulation-based inference (SBI) using neural posterior estimation (NPE) to tune an FE model of the human middle ear against subject-specific data. The training dataset consisted of 10,000 FE simulations of stapes velocity, ear-canal input impedance, and absorbance, paired with seven FE parameter values sampled within plausible ranges. By using simulated data, we generated a diverse training dataset, enabling efficient learning by the neural network (NN). The NN learned the association between parameters and simulation outcomes, providing a probability distribution of parameter values, which could be used to produce subject-specific computational inferences. By accounting for noise and test-retest variability, the method provided a probability distribution of parameters, rather than a single set, fitting three experimental datasets simultaneously. Importantly, examining the inferred parameter distributions alongside prior knowledge of normal ranges enables individualized differential inference used for diagnosis. SBI offers an objective alternative to sensitivity analyses, uncovering parameter interactions, supporting personalized diagnosis and treatment, and compensating for limited clinical training data. This method is applicable to any computational model, enhancing its potential for improved patient outcomes.
(© 2025. The Author(s).)

Declarations. Competing interests: The authors declare no competing interests.