eduzhai > Applied Sciences > Engineering >

High space-bandwidth in quantitative phase imaging using partially spatially coherent optical coherence microscopy and deep neural network

  • king
  • (0) Download
  • 20210505
  • Save

... pages left unread,continue reading

Document pages: 13 pages

Abstract: Quantitative phase microscopy (QPM) is a label-free technique that enables tomonitor morphological changes at subcellular level. The performance of the QPMsystem in terms of spatial sensitivity and resolution depends on the coherenceproperties of the light source and the numerical aperture (NA) of objectivelenses. Here, we propose high space-bandwidth QPM using partially spatiallycoherent optical coherence microscopy (PSC-OCM) assisted with deep neuralnetwork. The PSC source synthesized to improve the spatial sensitivity of thereconstructed phase map from the interferometric images. Further, compatiblegenerative adversarial network (GAN) is used and trained with pairedlow-resolution (LR) and high-resolution (HR) datasets acquired from PSC-OCMsystem. The training of the network is performed on two different types ofsamples i.e. mostly homogenous human red blood cells (RBC) and on highlyheterogenous macrophages. The performance is evaluated by predicting the HRimages from the datasets captured with low NA lens and compared with the actualHR phase images. An improvement of 9 times in space-bandwidth product isdemonstrated for both RBC and macrophages datasets. We believe that thePSC-OCM+GAN approach would be applicable in single-shot label free tissueimaging, disease classification and other high-resolution tomographyapplications by utilizing the longitudinal spatial coherence properties of thelight source.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×