Thickness Measurements from Single X-ray Phase-contrast Speckle Projection

We propose a one-shot thickness measurement method for sponge-like structures using a propagation-based X-ray phase-contrast imaging (P-PCI) method. In P-PCI, the air-material interface refracts the incident X-ray. Refracted many times along their paths by such a structure, incident X-rays propagate randomly within a small divergent angle range, resulting in a speckle pattern in the captured image. We found structure thickness and contrast of a phase-contrast projection are directly related in images. This relationship can be described by a natural logarithm equation. Thus, from the one phase-contrast view, depth information can be retrieved from its contrast. Our preliminary biological experiments indicate promise in its application to measurements requiring in vivo and ongoing assessment of lung tumor progression.

💡 Research Summary

The paper introduces a novel one‑shot thickness measurement technique for sponge‑like (highly porous) structures using propagation‑based X‑ray phase‑contrast imaging (P‑PCI). In P‑PCI, variations in refractive index at air‑material interfaces cause X‑rays to experience small angular deviations. When an X‑ray traverses a porous medium, it encounters a multitude of such interfaces; each encounter adds a tiny refraction, and the cumulative effect is a random walk within a narrow divergence cone. At the detector plane this random walk produces a speckle pattern whose statistical contrast reflects the number of refractions, i.e., the thickness of the material traversed.

The authors first demonstrate, through simulations and controlled phantom experiments, that the speckle contrast (defined as the standard deviation of intensity normalized by the mean) grows monotonically with sample thickness. Empirically, the relationship follows a natural‑logarithm law:

 C = a · ln(t) + b

where C is the measured contrast, t the physical thickness, and a, b are constants determined by experimental parameters such as X‑ray energy, propagation distance, detector resolution, and material composition. By calibrating a and b with a set of reference samples, the thickness of an unknown region can be extracted from a single phase‑contrast projection without any tomographic rotation or iterative reconstruction.

To assess biological relevance, the method was applied to ex‑vivo lung tissue containing growing tumor nodules. Lung parenchyma is intrinsically porous, making it an ideal test case. As the tumor expands, its local thickness increases, leading to a predictable rise in speckle contrast that adheres to the same logarithmic model. This proof‑of‑concept suggests that in‑vivo monitoring of tumor progression could be achieved with a minimal radiation dose and a single exposure, a significant advantage over conventional CT which requires multiple angles and higher dose.

Key advantages highlighted include: (1) rapid acquisition—only one projection is needed; (2) reduced radiation exposure compared with multi‑angle CT; (3) potential for real‑time monitoring in clinical settings; and (4) compatibility with existing P‑PCI setups. Limitations are also discussed: speckle contrast is sensitive to detector noise and requires high‑resolution, low‑noise sensors; the calibration constants a and b are experiment‑specific, necessitating routine calibration for different tissues or imaging geometries; and the method assumes a relatively uniform, isotropic porous structure, which may not hold for highly heterogeneous media.

The authors conclude that the logarithmic contrast‑thickness relationship provides a simple yet powerful tool for extracting depth information from a single phase‑contrast image. Future work will focus on extending the model to heterogeneous tissues, automating calibration procedures, and integrating the technique into clinical workflows for applications such as lung cancer surveillance, monitoring of pulmonary fibrosis, and other pathologies where tissue thickness changes are diagnostically important.