AI-Enhanced Microscopy Revolutionizes Real-Time Cellular Imaging
Engineers at the University of California San Diego have developed an AI-enhanced microscopy technique that allows for real-time, high-resolution video imaging of live cells. This advancement, published in Nature Communications, utilizes an algorithm that significantly speeds up the image processing of structured illumination microscopy (SIM), a method that enhances image detail by using patterned light. The new technique, called unrolled blind-SIM (UBSIM), integrates artificial intelligence to produce high-quality images instantly, avoiding the introduction of false details. This innovation allows scientists to observe cellular processes in real time, with the ability to capture rapid changes in structures like the endoplasmic reticulum at up to 50 frames per second.