[10a-A13-1~12] 3.3 Biomedical optics

9:30 AM - 9:45 AM JST(12:30 AM - 12:45 AM UTC)English Presentation

[10a-A13-3] Four-frame-based high-speed multiple-contrast dynamic optical coherence tomography enabled by deep learning and non-uniform-time scanning

〇(D)Yusong Liu¹, Ibrahim Abd El-Sadek², Rion Morishita¹, Chettanat Padungatthakij³, Atsuko Furukawa⁴, Satoshi Matsusaka⁴, Yoshiaki Yasuno¹ (1.COG, Univ. Tsukuba, 2.Damietta Univ., 3.KMITL, 4.Univ. Tsukuba)

Keywords:

dynamic optical coherence tomography,high-speed imaging,deep learning

We demonstrated a neural network-based method incorporating non-uniform temporal sampling to generate multi-contrast dynamic optical coherence tomography (MC-DOCT) images. Using only 1/8 OCT frames of conventional method, this technique enables high-speed MC-DOCT imaging (6 s/volume).