Remarks on propagation of discontinuities in stationary radiative transfer

Basic Information

• Paper ID: 2510.12130
• Title: Remarks on propagation of discontinuities in stationary radiative transfer
• Author: Daisuke Kawagoe (Graduate School of Informatics, Kyoto University)
• Classification: math.AP (Mathematical Analysis)
• Publication Date: October 15, 2025
• Paper Link: https://arxiv.org/abs/2510.12130

Abstract

This paper investigates the propagation of discontinuities in solutions to the stationary transport equation with incident boundary conditions. Under generalized convexity conditions, it is known that solutions exhibit only boundary-induced discontinuities, which originate from discontinuous boundary data, propagate along positive characteristic lines, and can be reconstructed from boundary measurements via the inverse X-ray transform. The paper observes that when generalized convexity conditions are violated, coefficient-induced discontinuities also appear—discontinuities in solutions caused by discontinuous coefficients. If the set of coefficient discontinuity points contains at most finitely many flat parts, coefficient-induced discontinuities do not affect the inverse X-ray transform. The article further demonstrates that under generalized convexity conditions, the three-dimensional inverse problem can be reduced to a two-dimensional problem, and presents numerical experiments.

Research Background and Motivation

Problem Background

The core problem studied in this paper concerns the stationary radiative transfer equation: $\xi \cdot \nabla_x I(x, \xi) + (\mu_a(x) + \mu_s(x))I(x, \xi) = \mu_s(x) \int_{S^{d-1}} p(x, \xi, \xi')I(x, \xi') d\sigma_{\xi'}$

This equation describes photon propagation in turbid media (such as biological tissue), where:

• $I(x, \xi)$: photon density at position $x$ in direction $\xi$
• $\mu_a$: absorption coefficient
• $\mu_s$: scattering coefficient
• $p$: scattering phase function
• $\mu_t = \mu_a + \mu_s$: total attenuation coefficient

Research Motivation

1. Theoretical Completeness: Existing theory primarily studies boundary-induced discontinuities under generalized convexity conditions, but this condition may not hold in practical applications
2. Application-Driven: Medical imaging techniques such as optical coherence tomography require reconstruction of internal parameters from boundary measurements
3. Numerical Inspiration: Numerical experiments in reference 4 show that successful reconstruction is possible even when generalized convexity conditions are violated
4. Inverse Problem Solving: Understanding discontinuity propagation is crucial for the stability of the inverse X-ray transform

Limitations of Existing Methods

• Strict dependence on generalized convexity conditions
• Consideration of only boundary-induced discontinuities
• Lack of systematic analysis of coefficient discontinuity effects
• High computational complexity in three-dimensional problems

Core Contributions

1. Extended Discontinuity Theory: Systematically analyzed the emergence and propagation of coefficient-induced discontinuities when generalized convexity conditions are violated
2. Established New Classification Framework: Classified solution discontinuities into boundary-induced and coefficient-induced categories
3. Proved Stability of Inverse Transform: Demonstrated that when the coefficient discontinuity set contains at most finitely many flat parts, coefficient-induced discontinuities do not affect the inverse X-ray transform
4. Proposed Dimension Reduction Method: Under generalized convexity conditions, reduced the three-dimensional inverse problem to a two-dimensional problem
5. Provided Numerical Verification: Validated theoretical results through concrete experiments

Detailed Methodology

Task Definition

Investigation of discontinuities in solutions to the boundary value problem (1.1)-(1.2) of the stationary transport equation:

• Input: Boundary condition $I_0(x,\xi)$, coefficients $\mu_a, \mu_s, p$
• Output: Discontinuity set of solution $I(x,\xi)$
• Objective: Reconstruct attenuation coefficient $\mu_t$ from boundary measurements

Theoretical Framework

1. Review of Generalized Convexity Condition

For domain partition $\{Ω_j\}_{j=1}^N$, the generalized convexity condition requires that for all $(x,ξ) \in Ω \times S^{d-1}$, the half-line $\{x-tξ|t≥0\}$ intersects $∂Ω_0$ at finitely many points.

2. Integral Representation of Solutions

Through integration along characteristic lines, the integral equation for the solution is obtained: $I(x,ξ) = \exp(-M_t(x,ξ;τ_-(x,ξ)))I_0(P(x,ξ),ξ) + \int_0^{τ_-(x,ξ)} \mu_s(x-sξ)\exp(-M_t(x,ξ;s)) \times \int_{S^{d-1}} p(x-sξ,ξ,ξ')I(x-sξ,ξ')dσ_{ξ'}ds$

where $M_t(x,ξ;s) = \int_0^s \mu_t(x-rξ)dr$.

3. Classification of Discontinuities

Two-Dimensional Case (Theorem 3.1): $\text{disc}(I) \subset \text{disc}_B \cup \text{disc}_C$

where:

• $\text{disc}_B$: boundary-induced discontinuities
• $\text{disc}_C = \cup_{j=1}^K \{(x+tξ_±^{(j)}, ξ_±^{(j)}) | x \in L_j, 0 ≤ t < τ_+(x,ξ_±^{(j)})\}$

Three-Dimensional Case: $\text{disc}_{C,3d} = \cup_{j=1}^K \{(x+tξ,ξ) | x \in F_j, ξ \in S_j^2, 0 ≤ t < τ_+(x,ξ)\}$

Technical Innovations

1. Iterative Decomposition Method: Decompose the solution as $I = \sum_{n=0}^∞ I^{(n)}$, where $I^{(0)}$ contains boundary data and $I^{(n≥1)}$ represents scattering contributions
2. Discontinuity Propagation Analysis:
   • Discontinuities in $I^{(0)}$ originate from boundary data and the $M_t$ function
   • Discontinuities in $I^{(n≥1)}$ originate solely from coefficients
3. Jump Calculation: For boundary-induced discontinuities, the jump is: $[I](x^*,ξ^*) = C\exp\left(-\int_0^{τ_-(x^*,ξ^*)} \mu_t(x^*-rξ^*)dr\right)$

Experimental Setup

Three-Dimensional Numerical Experiment Configuration

• Domain: Unit ball $Ω = \{x \in \mathbb{R}^3 | |x| < 1\}$
• Domain Partition:
  • $Ω_1$: annular region, $\mu_a = 0.3$
  • $Ω_2$: spherical region, $\mu_a = 0.2$
  • $Ω_3$: remaining region, $\mu_a = 0.1$
• Scattering Coefficient: $\mu_s = 0.3$ (constant)
• Phase Function: Henyey-Greenberg kernel, $g = 0.9$

Reconstruction Method

1. Generate discontinuities using boundary condition (2.3)
2. Observe X-ray transform through jump measurements
3. Apply A-analytic theory for stable reconstruction
4. Employ classical methods such as filtered backprojection

Experimental Results

Main Theoretical Results

Theorems 3.1-3.3 establish that:

1. Under finite flat parts condition, solution discontinuities are controllable
2. Jump formula (2.4) for boundary-induced discontinuities remains valid
3. Coefficient-induced discontinuities do not affect stability of the inverse X-ray transform

Numerical Reconstruction Results

• Figure 4.2: Reconstructed $\mu_t$ distribution image, clearly showing three distinct regions
• Figure 4.3: Cross-sectional plot along diagonal, quantitatively verifying reconstruction accuracy
• Reconstruction successfully recovered main features of the original coefficient distribution

Dimension Reduction Verification

Under generalized convexity conditions, by restricting directions to $S_0^2 = \{ξ \in S^2 | ξ_3 = 0\}$, successfully reduced the three-dimensional problem to a two-dimensional tomography problem.

Related Work

Foundational Theory

• Anikonov et al. 1: Established basic theoretical framework under generalized convexity conditions
• Chen and Kawagoe 5: Systematically studied propagation of boundary-induced discontinuities

Application Background

• Arridge 2,3: Theoretical foundations and applications of optical coherence tomography
• Fujiwara et al. 6: Fourier methods for inverse source problems

Numerical Methods

• Chen et al. 4: Numerical implementation of scattered signal tomography
• Natterer 8: Mathematical foundations of computed tomography

Conclusions and Discussion

Main Conclusions

1. Theoretical Extension: Successfully extended discontinuity theory to non-generalized-convexity cases
2. Stability Guarantee: Proved stability of inverse transform under finite flat parts condition
3. Computational Simplification: Provided effective dimension reduction from three to two dimensions
4. Practical Value: Provided theoretical support for applications such as optical coherence tomography

Limitations

1. Condition Restrictions: Still requires assumption of "finitely many flat parts"
2. Scattering Complexity: Analysis of complex scattering cases requires further development
3. Numerical Stability: Numerical stability in high dimensions requires further investigation
4. Practical Noise: Theoretical analysis insufficiently addresses measurement noise effects

Future Directions

1. Further relax restrictions on geometric conditions
2. Develop more stable numerical reconstruction algorithms
3. Extend to time-dependent transport equations
4. Incorporate machine learning methods to improve reconstruction efficiency

In-Depth Evaluation

Strengths

1. Theoretical Depth: Rigorous and thorough analysis of discontinuity propagation with complete mathematical derivations
2. Practical Value: Direct application to important fields such as medical imaging
3. Methodological Innovation: Original classification and analysis of coefficient-induced discontinuities
4. Completeness: Forms complete chain from theoretical analysis to numerical verification

Weaknesses

1. Assumption Conditions: Geometric assumptions on coefficient discontinuities remain relatively strong
2. Numerical Verification: Numerical experiments are relatively simple, lacking verification in complex scenarios
3. Noise Robustness: Insufficient discussion of noise effects in practical measurements
4. Computational Complexity: Insufficient analysis of algorithm complexity

Impact

1. Theoretical Contribution: Provides important theoretical tools for transport theory and inverse problems
2. Application Prospects: Broad application potential in optical coherence tomography, nuclear engineering, and other fields
3. Methodological Inspiration: Provides exemplar for handling discontinuities in similar partial differential equations
4. Foundation for Future Research: Establishes foundation for investigating more general cases

Applicable Scenarios

1. Optical Coherence Tomography: Tissue parameter reconstruction in medical diagnosis
2. Nuclear Reactor Physics: Material parameter identification in neutron transport
3. Atmospheric Science: Medium property inversion in radiative transfer
4. Geophysics: Subsurface structure reconstruction in seismic wave propagation

References

This paper primarily references the following important works:

1. Pioneering work by Anikonov et al. on generalized convexity conditions
2. Survey by Arridge on optical coherence tomography
3. Prior research by Chen and Kawagoe on boundary-induced discontinuities
4. Classical textbook by Natterer on computed tomography

Overall Assessment: This is a high-quality paper of significant theoretical value and practical importance in the field of mathematical analysis. The author successfully extended the theory of discontinuity propagation in stationary radiative transfer, providing a more general theoretical framework for solving inverse problems. While there is room for improvement in certain technical details and experimental verification, its theoretical contributions and methodological innovations make it an important advance in the field.