Light passing through a turbid medium like milk or living tissue scatters in all directions. Reconstructing where light was absorbed more strongly is like guessing the shape of a fallen object from a puddle on the asphalt. Optical tomography solves this problem, but usually requires large computational resources.
The new GS-DOT method represents the absorption map as a sum of blurred blobs — Gaussian primitives. Each blob has its own brightness, size, and orientation. The algorithm adjusts them so that the calculated light propagation matches the measured one. This is similar to how an artist builds a picture from many translucent brushstrokes.
The method was tested on synthetic tissue models. Even with added measurement noise, it accurately determines the position and strength of absorption. This approach speeds up data processing and makes tomography more accessible for clinics.
An artist with a few brushes now performs just as well as one with a thousand. Light again helps to look inside the body without incisions.
Gaussian blobs are mathematical bell-shaped functions. Their shape describes how light spreads during scattering, so they are ideal for modeling photon diffusion.
In the series 'Star Trek', the medical tricorder scans a patient in seconds. New reconstruction algorithms like GS-DOT bring such devices closer to reality, making signal processing fast and compact.