A group of researchers has created a deep learning-based ultrasound hologram generation framework technology that allows them to freely configure the form of focused ultrasound in real time using holograms. It is expected to be a foundational technology in brain stimulation and precision treatment.
Ultrasound can treat conditions such as Alzheimer’s, depression, and pain. The issue is that it is difficult to stimulate related brain areas selectively. A technology capable of freely focusing ultrasound on a desired area has been proposed to solve this problem, but it has limitations.
To overcome the limitations, the team proposed a deep learning-based learning framework to embody free and accurate ultrasound focusing in real-time by learning to generate ultrasound holograms. As a result, the team demonstrated that it was possible to focus ultrasound more precisely into the desired form in a hologram creation time close to real-time and up to 400 times faster than the existing ultrasound hologram generation algorithm method.
The research team’s proposed deep learning-based learning framework learns to generate ultrasonic holograms through self-supervised learning. The research team proposed a learning methodology for producing ultrasonic holograms, a deep learning network optimized for producing ultrasonic holograms, and a new loss function, all while demonstrating the validity and excellence of each component through simulations and actual experiments.
Related Content: Orbital Angular Momentum Multiplexing Holography