Post-signal-processing techniques of refocusing and digital aberration correction of optical coherence tomography (OCT) restore the spatial resolution deteriorated by optical aberrations. In the case of in vivo biological tissues with Fourierdomain OCT, an ultrafast volumetric acquisition is required to avoid motion distortion. In point-scanning OCT, a fast scan is required, and the bulk phase shifts among surrounding A-lines should be corrected. A low duty cycle might be necessary to obtain the consistent en face image with a high-speed raster scan. Recently, we have demonstrated a Lissajous-patternbased probe beam scanning and motion correction algorithm. In this study, we demonstrate a Lissajous-cycle-wise (LCW) digital refocus algorithm. The algorithm of LCW digital refocus is based on a convolution operation with an inverse filter. The complex OCT signals sampled with a Lissajous cycle are going to be convolved with an inverse filter of defocus. The reconstruction from Lissajous data to the Cartesian coordinate assigns a Cartesian grid to each A-line. By merging A-lines of different cycles, blurring due to defocus perpendicular to the scanning trajectory is mitigated by other cycle data. The phantom experiment is applied for the proof of concept. A prototype 1-μm spectral-domain OCT system is used for experiments. The A-line rate is 92 kHz. The blurred image of a phantom by defocusing is sharpened by the LCW digital refocus process along with all directions in the en face plane. Although the restored resolution does not reach the diffraction limit, an ultrafast volumetric acquisition is not required when one Lissajous cycle is significantly faster than sample motion.
We propose a new multi-focus average method for optical coherence tomography, to reduce the multiple scattering signals and improve the visibility of the sample structure in the deep region. It combines computational refocusing, complex averaging, and multiple acquisitions with focal shifting. A scattering phantom was measured to validate the suppression of multiple scattering signals and the contrast improvement at the deep region.
In this research, we demonstrated a CNN-based DOPU estimation algorithm without polarization-sensitive OCT signals. The CNN was trained with pairs of retinal OCT (input) and DOPU (teaching) images. The recall and precision of RPE abnormal appearances between true DOPU and synthesized DOPU of pathological eyes were calculated. For normal eyes, the grader evaluated the soundness of the RPE appearance for true DOPU and synthesized DOPU. The recalls are relatively good (0.74-0.95), while the precisions highly depend on the types of abnormalities (0.37-0.98). Five RPE abnormalities are found in synthesized DOPU within 25 synthesized DOPU B-scans while there is no abnormality in the true DOPU.
A three-dimensional label-free multi-contrast imaging for ex vivo tissue investigation is presented. Computational refocusing is implemented in a Jones-matrix polarization-sensitive optical coherence tomography (PS-OCT) system to overcome the trade-off between imaging depth and lateral resolution. The application of multiple contrast imaging, including intensity, birefringence, and degree-of-polarization uniformity (DOPU), is demonstrated by phantom, porcine muscle, and zebrafish measurements. Extended imaging depth with enhanced lateral resolution over millimeter is achieved. In tissue imaging we find some altered birefringence and DOPU estimation, whose size and alteration are proportional to defocus amount. This biased estimation can be numerically mitigated after applying computational refocusing.
We will present a deep convolutional neural network (DCNN) based estimators for optical coherence tomography (OCT). The DCNNs analyze local OCT speckle patterns and estimate the sample’s scatterer density and OCT resolutions. This estimator is intensity invariant, i.e., it does not use the net signal strength of OCT even to estimate the scatterer density. The DCNN is trained by a huge training dataset that was generated by a simple simulator of OCT imaging. This method is validated either by scattering phantom and in vitro tumor spheroid, and good accuracies of the estimation were shown.
We demonstrate volumetric phase contrast imaging by using optical coherence tomography (OCT). In general, the randomness of the scatterers’ distribution prohibits the volumetric measurement of a meaningful phase in a scattering mode. Our method uses complex numerical manipulation of an en-face complex OCT and gives a transversally differential phase image similar to a differential interference contrast microscope (DIC). Not like the DIC, our method can arbitrarily select the amount and direction of the shear after the OCT acquisition. In addition it provides DIC-like images at arbitrary depths. This method is validated by using a 840-nm spectral domain OCT system. A zebrafish sample is measured over a 1-mm × 1-mm transversal scanning range.
A multi-functional optical coherence microscopy capable of computational refocusing, tissue dynamics and birefringence imaging, and scatterer density estimation is demonstrated. It is applied to cell spheroid, ex vivo animal tissues.
Optical coherence tomography (OCT) has been widely used for imaging biological sample due to its capability of three-dimensional (3-D) reconstruction. Recently, polarization-sensitive optical coherence tomography (PSOCT) has been used to investigate polarization properties of samples such as retina and muscle [1,2]. Such extensions provide additional contrasts other than traditional reflected intensity and offer a 3-D multi-functional reconstruction of materials and biological tissue. PS-OCT uses A-scan wise computation to obtain the polarization properties from a set of OCT images. In this computation, the lateral resolution is implicitly assumed as infinitely high. However, the birefringence measurement itself will be affected by the lateral optical resolution, defocus, and aberrations. One simple way to obtain high-resolution birefringent data is to use an objective with higher NA. However, OCT also suffers from the trade-off between the lateral resolution and the depth-of-focus (DOF), which might limit its application in thick samples. There have been various methods to overcome this issue. Hardware solutions such as mechanical depth scanning [3] have been reported, but the additional configurations would increase the complexity of system. On the other hand, computational methods do not require additional hardware set-up and thus can be easily adopted, such as interferometric synthetic aperture microscopy (ISAM) [4] and forward model based computational refocusing [5]. However, to our acknowledgement, there are only few reports of long imaging depth for 3-D birefringence imaging. One presented work combined the ISAM with PS-OCT [6]. But the impact of refocusing on multi-contrast imaging, such as artifact in birefringence measurement, has not been thoroughly investigated. In this paper, we present a DOF extended polarization-sensitive imaging by applying computational refocusing to Jones-matrix based PS-OCT (JM-OCT). Computational refocusing is applied to each of four polarization channels (images) of JM-OCT, and the birefringence and degree-of-polarization-uniformity (DOPU) images are computed from the refocused OCT images. Enhanced lateral resolution away from the focus plane and hence extended DOF are demonstrated through phantom and ex vivo porcine muscle measurements. The ex vivo porcine muscle measurement also suggests that the refocusing may reduce the birefringence artifacts.
Deep convolutional neural network (CNN) based estimators for optical coherence tomography (OCT) are presented. The CNN analyze local OCT speckle patterns and estimate the sample’s scatterer density and OCT resolutions. This estimator is intensity invariant, i.e., it does not use the net signal strength of OCT even to estimate the scatterer density. The CNN is trained by a huge training dataset that was generated by a simple simulator of OCT imaging. And hence, the CNN is trained without experimental datasets. The performance of CNN was evaluated by numerically generated OCT images, and good accuracies of the estimation were shown.
We demonstrated computational multi-directional optical coherence tomography to visualize the microstructural directionality of a tissue. It uses numerical sub-aperture masks to the spatial frequency spectrum of en face OCT signals. An OCT signal consists of the intensity and phase, and they are processed independently. By applying the sub-aperture masks with several directions to the intensity signal’s spectrum, we can determine the dominant en face direction of the microstructure. In addition, sub-resolution depth-orientation of the microstructure is obtained by sub-aperture processing of the OCT phase signal. The microstructural directionality of bovine Achilles tendon and chicken breast muscle samples were visualized.
Bulk phase error (BPE) caused by environmental fluctuations will be a crucial problem for complex numerical processing of optical coherence tomography (OCT). We established a fully numerical phase stabilization method so-called smart-integration-path (SIP) method, which first estimates the vectorial gradient of BPE from a volumetric OCT data, then computes BPE from the gradient, and finally corrects BPE. The performance of SIP was subjectively evaluated by computational refocusing performance of ex vivo samples. The performance was also objectively evaluated by the standard deviations of spatial frequency spectra of en face OCT images. Both analyses showed positive effects of the BPE correction.
A numerical method to cancel the phase error caused by sample bulk motion from SD-OCT volume data is presented. This method first measures the vectorial gradient field of the phase error. The phase error is then estimated by path integration operation with non-standard integration paths. The performance of this method is validated by assessing the image quality of numerically refocused OCT images which is generated by complex holographic signal processing.
A convolutional neural networks (CNN) based scatterer density estimator for optical coherence tomography (OCT) is presented. In order to train the OCT, small patches of OCT speckle image were numerically generated. In this numerical image generation, the imaging parameters including the resolutions, probe power, signal-to-noise ratio, and scatterer density were randomly defined. So, the CNN was trained to estimate the imaging parameters from the generated OCT image patch. The results showed that our CNN estimator can estimate the parameters from the OCT speckle images.
Due to asynchronization between the acquisition trigger and K-clock trigger in a swept source optical coherence tomography (SS-OCT) system, trigger jitter causes the spectrum a temporal shift in the spectral domain and thus corrupts the measurement. We study ternary distribution of the jitter signal by measuring TiO2 phantom using a SS-OCT system, and it shows one-pixel spectral shift in the spectral domain.
A numerical method to cancel the phase error caused by sample bulk motion from SD-OCT volume data is presented. This method first measures the vectorial gradient field of the phase error. The phase error is then estimated by path integration operation with non-standard integration paths. The performance of this method is validated by assessing the image quality of numerically refocused OCT images which is generated by complex holographic signal processing.
A virtual multi-directional optical coherence tomography method to visualize directional microstructures of the tissue is demonstrated. The directional measurement is achieved by numerical aperture masks applied after the data acquisition. So multi-directional images are created only a single measurement with no hardware extension. By applying several types of aperture masks, we created not only multi-directional images but also multi-structural-frequency images. These images showed the difference appearance which come from the microstructural directionality of chicken breast muscle sample.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.