Developing deep learning based real-time ultrasonic hologram generation technology


Credit: IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Management (2022). DOI: 10.1109/TUFFC.2022.3219401

DGIST Division of Electrical Engineering and Pc Science Professor Hwang Jae-yoon’s crew developed a deep learning-based ultrasound hologram era framework know-how that may freely configure the type of centered ultrasound in actual time based mostly on holograms. It’s anticipated for use as a primary know-how within the area of mind stimulation and therapy that requires precision sooner or later.

Ultrasound is a protected know-how even used for prenatal examination. Since it might probably stimulate deep areas with out surgical procedure, ultrasound strategies for mind stimulation and therapy have just lately been studied. Earlier analysis has discovered that ultrasound mind stimulation can assist deal with circumstances, akin to Alzheimer’s illness, melancholy, and ache.

Nonetheless, the issue is that it’s troublesome to selectively stimulate associated areas of the mind during which a number of areas work together with one another on the similar time as a result of the present know-how focuses ultrasound right into a single small level or a big circle for stimulation. To unravel this drawback, a know-how able to freely focusing ultrasound on a desired space utilizing the hologram precept had been proposed, however has limitations, akin to low accuracy and lengthy calculation time to generate a hologram.

DGIST Professor Hwang Jae-yoon’s crew proposed a deep learning-based studying framework that may embody free and correct ultrasound focusing in actual time by studying to generate ultrasound holograms to beat the restrictions. In consequence, Professor Hwang’s crew demonstrated that it was doable to focus ultrasound into the specified kind extra precisely in a hologram creation time near actual time, and as much as 400 occasions sooner than the present ultrasound hologram era algorithm technique.

The deep learning-based studying framework proposed by the analysis crew learns to generate ultrasonic holograms by way of self-supervised studying. Self-supervised studying is a technique of studying to seek out the reply by discovering a rule by itself for information with no reply. The analysis crew proposed a technique for studying to generate ultrasonic hologram, a deep studying community optimized for ultrasonic hologram era, and a brand new loss perform, whereas proving the validity and excellence of every element by way of simulations and precise experiments.

DGIST Division of Electrical Engineering and Pc Science Professor Hwang Jae-yoon stated, “We applied deep learning technology to ultrasound holograms proposed relatively recently. As a result, we developed a technology that can freely, quickly and accurately generate and change the form of ultrasound beams,” and added, “We hope that the outcomes of this analysis are utilized in patient-specific precision brain stimulation know-how and common ultrasound fields (ultrasound imaging, thermal remedy, and so forth.).”

The paper is printed within the journal IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Management.

Extra data:
Moon Hwan Lee et al, Deep Studying-Primarily based Framework for Quick and Correct Acoustic Hologram Technology, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Management (2022). DOI: 10.1109/TUFFC.2022.3219401

Offered by
DGIST (Daegu Gyeongbuk Institute of Science and Expertise)

Creating deep studying based mostly real-time ultrasonic hologram era know-how (2022, December 29)
retrieved 29 December 2022

This doc is topic to copyright. Aside from any truthful dealing for the aim of personal examine or analysis, no
half could also be reproduced with out the written permission. The content material is supplied for data functions solely.

Click Here To Join Our Telegram Channel

Source link

When you have any considerations or complaints relating to this text, please tell us and the article shall be eliminated quickly. 

Raise A Concern