A Bayesian Generative Model With Gaussian Process Priors For Thermomechanical Analysis Of Micro-Resonators

Maximillian Fornitz Vording; Peter O. Okeyo; Juan José Rubio Guillamón; Mikkel Nørgaard Schmidt; Peter Emil Larsen; Tommy Sonne Alstrøm

doi:10.1109/MLSP.2019.8918876

A Bayesian Generative Model With Gaussian Process Priors For Thermomechanical Analysis Of Micro-Resonators

Maximillian Fornitz Vording, Peter O. Okeyo, Juan José Rubio Guillamón, Mikkel Nørgaard Schmidt, Peter Emil Larsen, Tommy Sonne Alstrøm

University of Copenhagen

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

179 Downloads (Pure)

Abstract

Thermal analysis using resonating micro-electromechanical systems shows great promise in characterizing materials in the early stages of research. Through thermal cycles and actuation using a piezoelectric speaker, the resonant behaviour of a model drug, theophylline monohydrate, is measured across the surface whilst using a laser-Doppler vibrometer for readout. Acquired is a sequence of spectra that are strongly correlated in time, temperature and spatial location of the readout.
Traditionally, each spectrum is analyzed individually to locate the resonance peak. We propose a Bayesian model using a warped Gaussian process prior taking the correlations into account and demonstrate on both synthetic and experimental data, that it yields better estimates of both location and amplitude of the resonance peak. Thus, the proposed model can give a more precise characterization of drugs, which is important in drug discovery and development

Original language	English
Title of host publication	Proceedings of 2019 IEEE International Workshop on Machine Learning for Signal Processing
Number of pages	6
Publisher	IEEE
Publication date	2019
ISBN (Print)	978-1-7281-0824-7
DOIs	https://doi.org/10.1109/MLSP.2019.8918876
Publication status	Published - 2019
Event	2019 IEEE 29th International Workshop on Machine Learning for Signal Processing - University of Pittsburgh, Pittsburgh, United States Duration: 13 Oct 2019 → 16 Oct 2019 Conference number: 29 https://ieeexplore.ieee.org/xpl/conhome/8911118/proceeding

Conference

Conference	2019 IEEE 29th International Workshop on Machine Learning for Signal Processing
Number	29
Location	University of Pittsburgh
Country/Territory	United States
City	Pittsburgh
Period	13/10/2019 → 16/10/2019
Internet address	https://ieeexplore.ieee.org/xpl/conhome/8911118/proceeding

Keywords

Bayesian learning and modeling
Gaussian processes
Drug characterisation
Thermomechanical analysis

Access to Document

10.1109/MLSP.2019.8918876

FulltextAccepted author manuscript, 5.03 MB

OpenUrl availability

Full text

Cite this

@inproceedings{3acf9d1172d64f07aafe78436e37246e,

title = "A Bayesian Generative Model With Gaussian Process Priors For Thermomechanical Analysis Of Micro-Resonators",

abstract = "Thermal analysis using resonating micro-electromechanical systems shows great promise in characterizing materials in the early stages of research. Through thermal cycles and actuation using a piezoelectric speaker, the resonant behaviour of a model drug, theophylline monohydrate, is measured across the surface whilst using a laser-Doppler vibrometer for readout. Acquired is a sequence of spectra that are strongly correlated in time, temperature and spatial location of the readout.Traditionally, each spectrum is analyzed individually to locate the resonance peak. We propose a Bayesian model using a warped Gaussian process prior taking the correlations into account and demonstrate on both synthetic and experimental data, that it yields better estimates of both location and amplitude of the resonance peak. Thus, the proposed model can give a more precise characterization of drugs, which is important in drug discovery and development",

keywords = "Bayesian learning and modeling, Gaussian processes, Drug characterisation, Thermomechanical analysis",

author = "Vording, {Maximillian Fornitz} and Okeyo, {Peter O.} and {Rubio Guillam{\'o}n}, {Juan Jos{\'e}} and Schmidt, {Mikkel N{\o}rgaard} and Larsen, {Peter Emil} and Alstr{\o}m, {Tommy Sonne}",

year = "2019",

doi = "10.1109/MLSP.2019.8918876",

language = "English",

isbn = "978-1-7281-0824-7",

booktitle = "Proceedings of 2019 IEEE International Workshop on Machine Learning for Signal Processing",

publisher = "IEEE",

address = "United States",

note = "2019 IEEE 29th International Workshop on Machine Learning for Signal Processing, MLSP 2019 ; Conference date: 13-10-2019 Through 16-10-2019",

url = "https://ieeexplore.ieee.org/xpl/conhome/8911118/proceeding",

}

Vording, MF, Okeyo, PO, Rubio Guillamón, JJ, Schmidt, MN, Larsen, PE & Alstrøm, TS 2019, A Bayesian Generative Model With Gaussian Process Priors For Thermomechanical Analysis Of Micro-Resonators. in Proceedings of 2019 IEEE International Workshop on Machine Learning for Signal Processing. IEEE, 2019 IEEE 29th International Workshop on Machine Learning for Signal Processing, Pittsburgh, Pennsylvania, United States, 13/10/2019. https://doi.org/10.1109/MLSP.2019.8918876

A Bayesian Generative Model With Gaussian Process Priors For Thermomechanical Analysis Of Micro-Resonators. / Vording, Maximillian Fornitz; Okeyo, Peter O.; Rubio Guillamón, Juan José et al.

Proceedings of 2019 IEEE International Workshop on Machine Learning for Signal Processing. IEEE, 2019.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - A Bayesian Generative Model With Gaussian Process Priors For Thermomechanical Analysis Of Micro-Resonators

AU - Vording, Maximillian Fornitz

AU - Okeyo, Peter O.

AU - Rubio Guillamón, Juan José

AU - Schmidt, Mikkel Nørgaard

AU - Larsen, Peter Emil

AU - Alstrøm, Tommy Sonne

N1 - Conference code: 29

PY - 2019

Y1 - 2019

N2 - Thermal analysis using resonating micro-electromechanical systems shows great promise in characterizing materials in the early stages of research. Through thermal cycles and actuation using a piezoelectric speaker, the resonant behaviour of a model drug, theophylline monohydrate, is measured across the surface whilst using a laser-Doppler vibrometer for readout. Acquired is a sequence of spectra that are strongly correlated in time, temperature and spatial location of the readout.Traditionally, each spectrum is analyzed individually to locate the resonance peak. We propose a Bayesian model using a warped Gaussian process prior taking the correlations into account and demonstrate on both synthetic and experimental data, that it yields better estimates of both location and amplitude of the resonance peak. Thus, the proposed model can give a more precise characterization of drugs, which is important in drug discovery and development

AB - Thermal analysis using resonating micro-electromechanical systems shows great promise in characterizing materials in the early stages of research. Through thermal cycles and actuation using a piezoelectric speaker, the resonant behaviour of a model drug, theophylline monohydrate, is measured across the surface whilst using a laser-Doppler vibrometer for readout. Acquired is a sequence of spectra that are strongly correlated in time, temperature and spatial location of the readout.Traditionally, each spectrum is analyzed individually to locate the resonance peak. We propose a Bayesian model using a warped Gaussian process prior taking the correlations into account and demonstrate on both synthetic and experimental data, that it yields better estimates of both location and amplitude of the resonance peak. Thus, the proposed model can give a more precise characterization of drugs, which is important in drug discovery and development

KW - Bayesian learning and modeling

KW - Gaussian processes

KW - Drug characterisation

KW - Thermomechanical analysis

U2 - 10.1109/MLSP.2019.8918876

DO - 10.1109/MLSP.2019.8918876

M3 - Article in proceedings

SN - 978-1-7281-0824-7

BT - Proceedings of 2019 IEEE International Workshop on Machine Learning for Signal Processing

PB - IEEE

T2 - 2019 IEEE 29th International Workshop on Machine Learning for Signal Processing

Y2 - 13 October 2019 through 16 October 2019

ER -

A Bayesian Generative Model With Gaussian Process Priors For Thermomechanical Analysis Of Micro-Resonators

Abstract

Conference

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this