COVID Test – Microscopy, AI Improve Accuracy

Researchers have paired microscopy with artificial intelligence to develop a fast, accurate, and cost-effective COVID test. While many methods currently exist to test for SARS-CoV-2, none use a label-free optical approach. So, the research team turned to a technique typically reserved for visualizing cells: spatial light image microscopy (SLIM), which facilitates chemical-free (or label-free) imaging.

SLIM images are blurry since the viruses are smaller than the diffraction limit. However, SLIM is highly sensitive, allowing the researchers to detect the viruses and differentiate between different types. They employed AI to determine the viruses based on SLIM data. It is possible to program an advanced deep neural network to recognize even the blurriest images with the proper training.

The researchers introduced the AI program to a pair of images: a stained SARS-CoV-2 particle producing fluorescence and a phase image captured with a fluorescence-SLIM multimodal microscope. They trained the AI to recognize these images as the same. With enough repetition, the machine detects the viruses directly from the SLIM, label-free images without the added support. The combination of SLIM and AI gives rise to an accurate COVID test.

Read more