Description
The reconstruction of three-dimensional (3D) distributions of scattering properties in diffuse media remains a challenge in biomedical imaging. There is considerable interest in improved optical imaging methods for breast cancer detection, as conventional X-ray mammography exposes patients to ionising radiation and may cause discomfort, discouraging regular screening. However, the highly scattering nature of biological tissue presents significant challenges for optical imaging techniques.
Our work demonstrates the potential of light in the Near-Infrared window (~600-900 nm) for imaging, using information from the scatter patterns obtained when laser light is directed through a sample.
Using this approach, we recorded changes in the scattering patterns of 633 nm and 855 nm light passing through a bovine gelatine sample containing an embedded resin target to accurately reconstruct it. Gelatine (clear, red and green) was used to simulate biological material with similar density but varying absorption properties. Scatter patterns recorded at different lateral, longitudinal and rotational coordinates provide the information required to construct the Radon transform (sinogram) of longitudinal slices through the vial. The inverse Radon transform is then applied to obtain a series of 2D cross-sectional tomographic reconstructions.
Initial results have demonstrated the reconstruction of objects embedded within several centimetres of gelatine tissue-mimicking material using 633 nm laser illumination. Current investigations are extending this approach to 855 nm, a wavelength anticipated to provide greater optical penetration depth.
These scans have yielded successful target reconstructions, with acquisition of the remaining datasets for a complete comparison ongoing. Current results indicate that, with further development, both red and infrared wavelengths within the NIR window could be utilised for detecting objects through centimetres of tissue-mimicking material.
| I am the presenting author | Yes |
|---|