In a new study published in Light Science & Application, researchers led by Professor Chulhong Kim have created a technique that uses artificial intelligence to overcome the problem of slow imaging speed.
FREMONT, Calif.: Thunder can be heard for a brief moment after a lightning strike. This is because the surrounding substance that has been struck by lightning absorbs the light, which is then converted into heat, causing the material to expand and emit sound. Photoacoustic Imaging (PAI), a technology that uses this phenomenon to take images from inside the body, is being tested as a new medical imaging device in a variety of preclinical and clinical settings. PAI technology uses the localization imaging approach, which involves the repeated imaging of the same area to produce extremely high spatial resolution beyond the physical limit, regardless of imaging depth. However, since many frames, each carrying the location target, must be overlapped to generate an adequately sampled high-density super-resolution image, this higher spatial resolution comes at the cost of temporal resolution. This made it difficult to use for research requiring immediate confirmation.
A team of scientists led by Professor Chulhong Kim and multi-institutional partners have created an AI-based localization PAI to overcome slow imaging speed issues in new work published in Light Science & Application. He was able to address all three of these concerns at once by applying deep learning to improve imaging speed and reduce the number of laser beams needed on the body. The study team was able to reduce the number of photos used in this procedure by more than tenfold and increase the imaging speed by twelvefold thanks to deep learning technologies. Localization photoacoustic microscopy and photoacoustic computed tomography times were cut in half, from 30 seconds to 2.5 seconds and from 30 minutes to 2.5 minutes, respectively.
This innovation opens the possibility of using PAI localization techniques in a variety of preclinical and clinical applications that require both high speed and exquisite spatial resolution, such as investigations of drug and hemodynamic responses. Above all, one of the most notable advantages of this technique is that it significantly reduces both the exposure to the laser beam of the living body and the imaging time, which provides relief to patients.