Robust estimation of skin physiological parameters from hyperspectral images using Bayesian neural networks

Manojlović, Teo (author), Tomanič, Tadej (author), Štajduhar, Ivan (author), Milanič, Matija (author)

Significance: Machine learning models for the direct extraction of tissue parameters from hyperspectral images have been extensively researched recently, as they represent a faster alternative to the well-known iterative methods such as inverse Monte Carlo and inverse adding-doubling (IAD). Aim: We aim to develop a Bayesian neural network model for robust prediction of physiological parameters from hyperspectral images. Approach: We propose a two-component system for extracting physiological parameters from hyperspectral images. First, our system models the relationship between the measured spectra and the tissue parameters as a distribution rather than a point estimate and is thus able to generate multiple possible solutions. Second, the proposed tissue parameters are then refined using the neural network that approximates the biological tissue model. Results: The proposed model was tested on simulated and in vivo data. It outperformed current models with an overall mean absolute error of 0.0141 and can be used as a faster alternative to the IAD algorithm. Conclusions: Results suggest that Bayesian neural networks coupled with the approximation of a biological tissue model can be used to reliably and accurately extract tissue properties from hyperspectral images on the fly.

• Language: English
• Keywords: medical physics, hyperspectral imaging, neural networks
• Work type: Article
• Typology: 1.01 - Original Scientific Article
• Organization: FMF - Faculty of Mathematics and Physics
• Publication status: Published
• Publication version: Version of Record
• Year: 2025
• Number of pages: Str. 016004-1-016004-17
• Numbering: Vol. 30, iss. 1
• PID: 20.500.12556/RUL-166647
• UDC: 616-073
• ISSN on article: 1083-3668
• DOI: 10.1117/1.JBO.30.1.016004
• COBISS.SI-ID: 223069699
• Publication date in RUL: 20.01.2025

Record is a part of a journal

• Title: Journal of biomedical optics
• Shortened title: J. biomed. opt.
• Publisher: SPIE--the International Society for Optical Engineering, International Biomedical Optics Society
• ISSN: 1083-3668
• COBISS.SI-ID: 18188071

Licences

• License: CC BY 4.0, Creative Commons Attribution 4.0 International
• Link: http://creativecommons.org/licenses/by/4.0/
• Description: This is the standard Creative Commons license that gives others maximum freedom to do what they want with the work as long as they credit the author.

Secondary language

• Language: Slovenian
• Keywords: medicinska fizika, hiperspektralno slikanje, nevronske mreže