[Asia Economy Reporter Hyunseok Yoo] Neurophyt, a brain disease AI solution specialist company, announced on the 9th that it has obtained Class 2 medical device certification from the Ministry of Food and Drug Safety for its brain imaging analysis software ‘Neurophyt Aqua (NEUROPHET AQUA) 2.0’.
‘Neurophyt Aqua’ segments and analyzes brain magnetic resonance imaging (MRI), reconstructs it into a three-dimensional model, and measures volume to indicate the degree of brain atrophy. Based on the analysis results, it produces an objective report that allows the assessment of brain atrophy levels that are difficult to distinguish with the naked eye.
The newly certified ‘Neurophyt Aqua 2.0’ uses a convolutional neural network (CNN)-based algorithm to automatically segment brain images, enabling more accurate brain disease analysis. The convolutional neural network is a deep learning technique used for image recognition, modeled after the structure of the human retina.
Neurophyt’s self-developed deep learning-based algorithm demonstrated a Dice coefficient of 0.89 for multi-brain region segmentation in an SCIE-level international journal last December, proving a higher accuracy than international standards.
This means that the agreement rate between the ground truth images created by imaging specialists and the images generated by the algorithm reaches 90%, a level at which differences between the two images are imperceptible to the human eye.
Kim Donghyun, co-founder and Chief Technology Officer (CTO) of Neurophyt, said, "Through this certification from the Ministry of Food and Drug Safety, the safety and excellence of AI medical devices have been objectively proven," adding, "We will continue to upgrade our products to provide brain imaging analysis indicators that both doctors and patients can trust." He also added, "We plan to start clinical trials within this year for Class 3 medical device approval."
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.
