See you in two years!
11-16 April 2026
Conference 13006 > Paper 13006-107
Paper 13006-107

Fast and accurate skin parameter estimation from hyperspectral images using random Fourier features

10 April 2024 • 13:40 - 14:00 CEST | Londres 1/Salon 8, Niveau/Level 0

Abstract

Hyperspectral imaging (HSI) is a powerful tool for noninvasive assessment of skin properties, as it can capture the spectral signatures of different skin layers and components. However, HSI also requires efficient and accurate methods for estimating skin parameters, such as the thickness, scattering, and absorption coefficients of each skin layer, from the measured spectra. In recent years, much research has been done regarding the use of machine learning (ML) methods for reducing the time and computational cost required for estimating parameters, compared to classical methods, such as the inverse Monte Carlo (IMC) or the inverse adding-doubling (IAD) algorithm. In this study, we investigated the impact of using random Fourier features (RFF) with a simple linear regression model, as well as with an artificial neural network (ANN), to estimate parameter values directly from the spectra. We compared the proposed models with the ANN and a 1D convolutional neural network (CNN), both trained using the raw spectra as input. All models were trained on simulated data and evaluated on both simulated and in vivo measured spectra using mean absolute error (MAE). We found that even simple linear regression with RFFs performs comparably to the neural networks trained on raw spectra while having much lower training and inference time. The best results were attained with the RFF-based ANN, having an overall MAE of 0.0226, which is an improvement compared to the 1D-CNN, having an MAE of 0.0284.

Presenter

Matija Milanic
University of Ljubljana, Faculty of mathematics and physics (Slovenia), Jozef Stefan Instutute (Slovenia)
Dr. Milanic is an associate professor in medical physics at the University of Ljubljana and a researcher at Jozef Stefan Institute. He studies how light interacts with tissues and how this knowledge can be used in clinical settings. He is interested in applying light's spectral properties to diagnose various conditions (such as skin lesions, joint inflammation, and peritonitis) and reveal tissue structure and function (such as oxygenation, scattering, and chromophore distribution). He also investigates how laser light affects biological tissues and how to improve laser treatments (such as hair removal, fat reduction, and skin rejuvenation).
Application tracks: AI/ML
Author
Faculty of Engineering, University of Rijeka (Croatia)
Author
University of Ljubljana (Slovenia)
Author
Ivan Stajduhar
Faculty of Engineering (Croatia)
Presenter/Author
Matija Milanic
University of Ljubljana, Faculty of mathematics and physics (Slovenia), Jozef Stefan Instutute (Slovenia)