(주)토모큐브, KAIST / 류동훈, 류동민, 민현석*, 박용근*
Abstract
Optical diffraction tomography measures the three-dimensional refractive index map of a specimen and visualizes biochemical phenomena at the nanoscale in a non-destructive manner. One major drawback of optical diffraction tomography is poor axial resolution due to limited access to the three-dimensional optical transfer function. This missing cone problem has been addressed through regularization algorithms that use a priori information, such as non-negativity and sample smoothness. However, the iterative nature of these algorithms and their parameter dependency make real-time visualization impossible. In this article, we propose and experimentally demonstrate a deep neural network, which we term DeepRegularizer, that rapidly improves the resolution of a three-dimensional refractive index map. Trained with pairs of datasets (a raw refractive index tomogram and a resolution-enhanced refractive index tomogram via the iterative total variation algorithm), the three-dimensional U-net-based convolutional neural network learns a transformation between the two tomogram domains. The feasibility and generalizability of our network are demonstrated using bacterial cells and a human leukaemic cell line, and by validating the model across different samples. DeepRegularizer offers more than an order of magnitude faster regularization performance compared to the conventional iterative method. We envision that the proposed data-driven approach can bypass the high time complexity of various image reconstructions in other imaging modalities.
(a) 에서 도식화된 것과 같이 ODT 영상이 제한적인 입사각에서만 촬영되었을 경우, (b) 좌측 (Raw 3D RI) 하단에서 볼 수 있듯이 정보가 없는 주파수 영역을 확인할 수 있다. 제안하는 방식 (Deep learning output)과 기존 복원 알고리즘의 결과 (Iterative TV), 대부분의 정보가 채워진 것을 확인할 수 있는데, 제안하는 방법은 기존 알고리즘에 비해 훨씬 짧은 계산 시간을 보여준다.
DongHun Ryu, Dongmin Ryu, YoonSeok Baek, Hyungjoo Cho, Geon Kim, Young Seo Kim, Yongki Lee, Yoosik Kim, Jong Chul Ye, Hyun-Seok Min, YongKeun Park