Fourier ptychographic microscopy (FPM) is a promising computational imaging technique, but conventional full-color digital pathology based on FPM is time-consuming.

Inspired by color matching, Associate Prof. PAN An, Prof. YAO Baoli, and Prof. MA Caiwen from the Xi'an Institute of Optics and Precision Mechanics (XIOPM) of the Chinese Academy of Sciences (CAS) reported a colorization method via color transfer FMP termed CFPM. The study was published in Science China-Physics Mechanics & Astronomy.

The CFPM, with the reconstruction time reducing by 2/3 and the precision sacrificed by only 0.4%, marks a great leap for the efficiency of FPM colorization compared with traditional methods. It can also be regarded as an "unsupervised transfer learning" based on physical models without iterative optimization in contrast to traditional transfer learning.

Technically, it is difficult to ensure the authenticity of color and the correctness in the display process, and it is hard to ensure the accuracy of color transfer while improving the efficiency.   

Therefore, the researchers established the mapping relationship between the CIE-XYZ color space and different FPM-based color spaces.  They compared different color transfer schemes and showed that using a low-resolution color image with the same field of view as the donor image was the best choice.   

"By transferring low-resolution true color texture information of optical microscopes to electronic microscopes, this method also enlightens us that we may dye true color for black-and-white images of an electronic microscope," said PAN An, corresponding author of the study.

Results of stained resting sporangia. (a, a1) LR donor image with the entire FOV of a 4× / 0.1NA objective and its close-up. (b, b1) FPM recovery image under green channel (515.0 nm) and its close-up. (c) Ground truth captured by a 10× / 0.3NA objective. (d) Staining results via CFPM. (Image by XIOPM)