Bioimaging boost with new algorithm
Image: sCMOS noise-correction algorithm generates artifact-free microscopy images.
A new algorithm developed by US researchers expands the use of advanced cameras for biological microscopy.
Pioneered by Professor Fang Huang from Purdue University’s Weldon School of Biomedical Engineering, the noise correction algorithm allows researchers to use scientific complementary metal-oxide semiconductor (sCMOS) cameras for a wide range of biological research.
While sCMOS cameras have provided significant advances in imaging speed, sensitivity and field of view compared with traditional detectors, the sensor introduces pixel-dependent noise that if left uncorrected, generates imaging artifacts and biases in quantification.
“When you are trying to use this for biological studies, it's very difficult to determine whether this fluctuation comes from the sample (photons) or from the camera itself,” explains Huang's colleague, Dr Sheng Liu.
Before (top) and after use of the new algorithm. [Purdue University/Sheng Liu and Fang Huang]
With this in mind, the researchers developed the new algorithm that corrects the noise, exploiting a general property of imaging systems, the optical transfer function.
"Based on our knowledge of how each of the 4 million pixels on our camera chip behave, we are able to estimate the actual photon level at each pixel location... and obtain a noise-corrected image," says Huang.
"Because our algorithm combines the noise and the likelihood for minimisation, it minimises the noise fluctuation while maintaining the underlying expected photon count and resolution of the image," he adds.
Crucially, the developed algorithm can generally be applied to sCMOS-based detection and quantitative analysis in a broad spectrum of microscopy techniques, such as light-sheet microscopy, total internal reflection fluorescence microscopy, fluorescence resonance energy transfer microscopy, speckle microscopy, and conventional fluorescence imaging.
What's more, as Huang points out, the fundamental principle can be applied to other fields where a maximum cutoff frequency exists, such as astronomy and photonics.
"We hope that these fields can now benefit from the increased quantum efficiency, field of view and imaging speed of sCMOS cameras without compromising its quantitative detection," he says.
The researchers have filed a patent application on the algorithm through the Purdue Research Foundation’s Office of Technology Commercialization.
Research is published in Nature Methods.