Data-efficient extraction of optical properties from 3D Monte Carlo TPSFs using Bi-LSTM transfer learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Aghili, Joubine, Imbach, Rémi, Pallarès, Anne, Schmitt, Philippe, Uhring, Wilfried
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908960051691520
author Aghili, Joubine
Imbach, Rémi
Pallarès, Anne
Schmitt, Philippe
Uhring, Wilfried
author_facet Aghili, Joubine
Imbach, Rémi
Pallarès, Anne
Schmitt, Philippe
Uhring, Wilfried
contents Time-Resolved Spectroscopy (TRS) is a powerful modality for non-invasive characterization of turbid media. However, extracting optical properties, absorption $μ_a$ and reduced scattering $μ_s'$, from 3D stochastic measurements remains computationally expensive for real-time applications. In this paper, we propose a data-efficient, physics-informed transfer learning strategy using a Bidirectional Long Short-Term Memory (Bi-LSTM) network. By leveraging a fast deterministic solver to establish a physical prior before fine-tuning on a restricted set of 3D Monte Carlo simulations, our model successfully bridges the analytical-to-stochastic domain gap. The proposed method eliminates the systematic bias of analytical models while maintaining a competitive error with near-instantaneous inference time.
format Preprint
id arxiv_https___arxiv_org_abs_2604_11437
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Data-efficient extraction of optical properties from 3D Monte Carlo TPSFs using Bi-LSTM transfer learning
Aghili, Joubine
Imbach, Rémi
Pallarès, Anne
Schmitt, Philippe
Uhring, Wilfried
Numerical Analysis
Computational Physics
Time-Resolved Spectroscopy (TRS) is a powerful modality for non-invasive characterization of turbid media. However, extracting optical properties, absorption $μ_a$ and reduced scattering $μ_s'$, from 3D stochastic measurements remains computationally expensive for real-time applications. In this paper, we propose a data-efficient, physics-informed transfer learning strategy using a Bidirectional Long Short-Term Memory (Bi-LSTM) network. By leveraging a fast deterministic solver to establish a physical prior before fine-tuning on a restricted set of 3D Monte Carlo simulations, our model successfully bridges the analytical-to-stochastic domain gap. The proposed method eliminates the systematic bias of analytical models while maintaining a competitive error with near-instantaneous inference time.
title Data-efficient extraction of optical properties from 3D Monte Carlo TPSFs using Bi-LSTM transfer learning
topic Numerical Analysis
Computational Physics
url https://arxiv.org/abs/2604.11437