Spectrally encoded endoscopy (SEE) is a minimally invasive optical imaging modality capable of fast confocal imaging of internal tissue structures. Modern SEE systems use coherent sources to image deep within the tissue and data are processed similar to optical coherence tomography (OCT); however, standard processing of SEE data via the Fast Fourier Transform (FFT) leads to degradation of the axial resolution as the bandwidth of the source shrinks, resulting in a well-known trade-off between speed and axial resolution. Recognizing the limitation of FFT as a general spectral estimation algorithm to only take into account samples collected by the detector, in this work we investigate alternative high-resolution spectral estimation algorithms that exploit information such as sparsity and the general region position of the bulk sample to improve the axial resolution of processed SEE data. We validate the performance of these algorithms using bothMATLAB simulations and analysis of experimental results generated from a home-built OCT system to simulate an SEE system with variable scan rates. Our results open a new door towards using non-FFT algorithms to generate higher quality (i.e., higher resolution) SEE images at correspondingly fast scan rates, resulting in systems that are more accurate and more comfortable for patients due to the reduced image time.
Telemedicine is an emerging technology that aims to provide clinical healthcare at a distance. Among its goals, the
transfer of diagnostic images over telecommunication channels has been quite appealing to the medical community.
When viewed as an adjunct to biomedical device hardware, one highly important consideration aside from the transfer
rate and speed is the accuracy of the reconstructed image at the receiver end. Although optical coherence tomography
(OCT) is an established imaging technique that is ripe for telemedicine, the effects of OCT data compression, which may
be necessary on certain telemedicine platforms, have not received much attention in the literature. We investigate the
performance and efficiency of several lossless and lossy compression techniques for OCT data and characterize their
effectiveness with respect to achievable compression ratio, compression rate and preservation of image quality. We
examine the effects of compression in the interferogram vs. A-scan domain as assessed with various objective and