blood flow and depth measurement method based on multi-exposure laser speckle contrast imaging(MELSCI) with 3D-CNN model

Le résumé fourni par la source

Laser speckle contrast imaging(LSCI) has been developed to measure blood perfusion non-invasively for a long time. However, there are some limitations to analyzing the random speckle phenomenon and relying on the statistical description in bio-application. This study aimed to verify the three-dimensional convolution neural network(3D-CNN) model for analyzing laser speckle images and predicting the perfusion velocity. The dataset for training deep learning was processed in the form of 3D-image and the image was from a real-time LSCI system. The model can potentially measure static and dynamic speckle information and predict perfusion velocity under the static tissue.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.