TY - JOUR
T1 - Phun-Net
T2 - Ready-to-use neural network for unwrapping quantitative phase images of biological cells
AU - Dardikman-Yoffe, Gili
AU - Roitshtain, Darina
AU - Mirsky, Simcha K.
AU - Turko, Nir A.
AU - Habaza, Mor
AU - Shaked, Natan T.
N1 - Publisher Copyright:
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
PY - 2020/2/1
Y1 - 2020/2/1
N2 - We present a deep-learning approach for solving the problem of 2π phase ambiguities in two-dimensional quantitative phase maps of biological cells, using a multi-layer encoder-decoder residual convolutional neural network. We test the trained network, PhUn-Net, on various types of biological cells, captured with various interferometric setups, as well as on simulated phantoms. These tests demonstrate the robustness and generality of the network, even for cells of different morphologies or different illumination conditions than PhUn-Net has been trained on. In this paper, for the first time, we make the trained network publicly available in a global format, such that it can be easily deployed on every platform, to yield fast and robust phase unwrapping, not requiring prior knowledge or complex implementation. By this, we expect our phase unwrapping approach to be widely used, substituting conventional and more time-consuming phase unwrapping algorithms.
AB - We present a deep-learning approach for solving the problem of 2π phase ambiguities in two-dimensional quantitative phase maps of biological cells, using a multi-layer encoder-decoder residual convolutional neural network. We test the trained network, PhUn-Net, on various types of biological cells, captured with various interferometric setups, as well as on simulated phantoms. These tests demonstrate the robustness and generality of the network, even for cells of different morphologies or different illumination conditions than PhUn-Net has been trained on. In this paper, for the first time, we make the trained network publicly available in a global format, such that it can be easily deployed on every platform, to yield fast and robust phase unwrapping, not requiring prior knowledge or complex implementation. By this, we expect our phase unwrapping approach to be widely used, substituting conventional and more time-consuming phase unwrapping algorithms.
UR - http://www.scopus.com/inward/record.url?scp=85079011463&partnerID=8YFLogxK
U2 - 10.1364/BOE.379533
DO - 10.1364/BOE.379533
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
C2 - 32206402
AN - SCOPUS:85079011463
VL - 11
SP - 1107
EP - 1121
JO - Biomedical Optics Express
JF - Biomedical Optics Express
SN - 2156-7085
IS - 2
ER -