Papers
Topics
Authors
Recent
Search
2000 character limit reached

Quantitative Multi-Modal Optical Coherence Photoacoustic Elastography

Published 17 Jun 2026 in physics.med-ph and math.NA | (2606.18990v1)

Abstract: We present a novel multi-modal optical coherence photoacoustic elastography (OCPE) framework, which combines two imaging modalities, optical coherence tomography (OCT) and photoacoustic tomography (PAT), to enable complementary absorption-scattering measurements for the extraction of quantitative tissue features via quasi-static elastography. For this, we develop a sophisticated hybrid inversion algorithm for merging the complementary information layers contained in both OCT and PAT-based elastography measurements, and perform systematic evaluations to assess the impact of hybrid elastography data on strain and stiffness reconstructions. Studies on a silicone elastomer phantom demonstrate that the combined OCT-PAT approach outperforms single-modality OCT elastography and PAT elastography, yielding higher strain signal-to-noise ratio and improved stiffness estimates. These results establish the advantage of multi-modal complementary imaging and data merging for accurate, high-resolution elastographic strain and stiffness mapping in both scattering and absorbing materials.

Summary

  • The paper presents a novel OCPE framework integrating OCT and PAT with a hybrid inversion algorithm, yielding precise strain and stiffness maps.
  • The methodology leverages dual-layer imaging fusion, feature tracking, and regularization to mitigate artefacts and resolve micrometer-scale tissue details.
  • Results show superior performance over single modalities, with accurate Young’s modulus and strain estimations critical for early cancer detection.

Quantitative Multi-Modal Optical Coherence Photoacoustic Elastography: An Expert Summary

Introduction and Motivation

This paper introduces a novel framework for quantitative tissue characterization, termed Optical Coherence Photoacoustic Elastography (OCPE), which integrates Optical Coherence Tomography (OCT) and Photoacoustic Tomography (PAT) to provide high-resolution, high-contrast, quantitative strain and stiffness mapping via quasi-static elastography (2606.18990). Conventional elastography methods based on ultrasound or MRI offer organ-level assessment but lack micrometer-scale resolution, which is crucial for detecting subclinical changes (e.g., early cancer or microstructural inhomogeneities). OCT and PAT, although promising individually, exhibit complementary strengths: OCT delineates scattering structures but suffers from shadow artefacts, whereas PAT visualizes absorption-dominated features but lacks comprehensive coverage in sparsely absorbing biological tissues. The OCPE approach is designed to overcome these limitations by merging the information content from both modalities using a sophisticated hybrid inversion algorithm.

Hybrid Imaging and Data Fusion Methodology

The multi-layered hybrid inversion algorithm for OCPE leverages feature tracking, segmentation, optical flow strategies, and regularized inverse parameter estimation. Unlike prior "multi-modal elastography" work, which treats elastography as an ancillary layer over imaging data, this framework genuinely fuses dual-modality elastography measurements at the reconstruction level. The algorithmic pipeline is adaptable, enabling both axial and lateral displacement estimation, and mitigates artefacts inherent in single modalities.

The core element is the developed data-fusing elastographic optical flow method (DEOFM). This method exploits dual-layered OCT and PAT data, segmenting the update estimation into modality-dependent steps to circumvent artefacts (e.g., shadowing, visibility gaps). The pipeline supports prior incorporation (e.g., feature tracking of titanium dioxide scatterers embedded in phantoms) and regularization strategies, yielding robust displacement and strain maps.

Quantitative Elastography: Inverse Problem Framework

Quantitative elastography in OCPE is articulated as an inverse problem with internal data. Two main regularized approaches are considered: the two-step sequential inversion—displacement/strain estimation followed by material parameter inversion—and the one-step direct parameter inversion from imaging data. The paper implements:

  • Direct Strain Inversion (DSI): Assumes uniform stress under compression, computing Young's modulus pixel-wise from the strain map. The formulation is E(x)=σ/ε(x)E(x) = \sigma / \varepsilon(x).
  • Nonlinear Landweber Iteration (NLI): Employs displacement maps to iteratively estimate Lamé parameters, leveraging the Fréchet derivative and its adjoint for regularized parameter updates.
  • Intensity-Based Inversion Method (IIM): Directly minimizes the difference between deformed and reference imaging intensities via piecewise-constant parameterization, robust to noise and dimensionality constraints.

All reconstruction methods are highly sensitive to the quality and information density of the fused imaging data, with frequent regularization and region-of-interest segmentation to suppress artefacts and ensure robust parameter identification.

Experimental Framework and Implementation

The combined OCT-PAT device is engineered to facilitate synchronized force-controlled loading, imaging, and force sensing. The phantom materials—silicone elastomer with titanium dioxide scatterers—mimic tissue scattering and absorption, with inclusions representing vascular mimics. The experimental design ensures quasi-static deformation, multi-stage loading (100–325 mN), and rigorous mechanical equilibrium prior to acquisition, enabling ground truth comparison via tensile testing.

Rigorous co-registration, segmentation, and intensity compensation procedures are deployed. Feature tracking is achieved using custom algorithms tailored for speckle distributions, and lateral/axial displacement is evaluated in both 2D and sparse 3D modes.

Results: Numerical Performance and Artefact Mitigation

Strain Mapping

OCPE consistently outperforms single-modality OCE and PAE in both axial and lateral strain reconstruction, demonstrating enhanced strain signal-to-noise ratios and higher fidelity to inclusion morphology. The DEOFM approach reduces artefact propagation, particularly shadowing effects in OCT and reconstruction artefacts in PAT. Inclusion regions are accurately localized; lateral constraint effects imposed by stiff inclusions are faithfully reflected in the strain fields.

Stiffness Estimation

Strong numerical results are reported in mean Young's modulus reconstructions. The intensity-based IIM achieves the closest match to ground truth values, with minimal underestimation for both inclusions and background. DSI underestimates inclusion stiffness due to the invalid assumption of homogeneous stress in the presence of mechanical inhomogeneity, while NLI overestimates background values and underestimates inclusions, reflecting sensitivity to strain input quality. The granular error analysis reveals that stiffness estimates depend significantly on inclusion orientation, imaging projection, and proximity to compression plates.

Notably, both DSI and NLI fail in single-modality datasets, underscoring the necessity for complementary OCT-PAT data. The regularization and segmentation strategies embedded in the hybrid inversion pipeline are critical for robust parameter mapping.

Practical and Theoretical Implications

The paper establishes OCPE as a superior modality for quantitative, high-resolution elastography in scattering and absorbing materials. On the practical side, OCPE is directly relevant for early cancer detection, precision tissue characterization, and in situ biopsy assessment, offering diagnostic visibility at scales inaccessible to US or MRI-based elastography. Theoretical implications center on the improved well-posedness of coupled-physics inverse problems, leveraging internal data for enhanced identifiability and stability.

Moreover, the hybrid inversion paradigm sets a precedent for future algorithmic integration across disparate imaging modalities, indicating that similar strategies could be employed in other coupled-physics domains (e.g., magneto-acoustic or opto-electronic imaging). The error analyses and uncertainty quantification underscore the importance of orientation and boundary effects, motivating further development in adaptive, multi-angle acquisition protocols and extensions to complex geometries.

Conclusions

The OCPE framework combines the strengths of OCT and PAT through rigorous multi-layered hybrid inversion, producing superior quantitative elastography results with high information density, robust artefact suppression, and strong numerical fidelity. The approach demonstrates clear advantages over single-modality methods, particularly in resolving stiffness inhomogeneities and mitigating artefacts. Future developments will likely target algorithmic refinement, in vivo adaptation, and expansion to multi-angle, multi-deformation acquisitions for further enhancement of diagnostic resolution and parameter uniqueness.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.