To enter the world of the fantastically small via microscopy, the main currency is either a ray of light or electrons. Strong beams, which yield clearer images, are damaging to specimens. On the other hand, weak beams can give noisy, low-resolution images.
In a new study, researchers at Texas A&M University unveiled a machine-learning-based algorithm that can reduce graininess in low-resolution images and reveal details that were otherwise buried within the noise.
“This can be extremely valuable for a myriad of applications, including clinical ones, like estimating the stage of cancer progression and distinguishing between cell types for disease prognosis.”
In conventional deep-learning-based image processing techniques, the number and network between layers decide how many pixels in the input image contribute to the value of a single pixel in the output image. This value is immutable after the deep-learning algorithm has been trained. However, fixing the number for the input pixels limits the performance of the algorithm.
To overcome this hurdle, Dr. Shuiwang Ji and his students developed another deep-learning algorithm that can dynamically change the size of the receptive field. Unlike earlier algorithms that can only aggregate information from a small number of pixels, their algorithm can pool from a larger area of the image if required.
“Deep-learning algorithms such as ours will allow us to potentially transcend the physical limit posed by light that was not possible before,” said Ji. “This can be extremely valuable for a myriad of applications, including clinical ones, like estimating the stage of cancer progression and distinguishing between cell types for disease prognosis.”