Dynamic micro-optical coherence tomography (DµOCT) is a technology that is capable of interrogating intracellular dynamics in intact, viable tissues. Towards our goal of advancing DµOCT for phenotyping cells, we imaged freshly excised human biopsies and performed correlative studies with histological results. To date, more than 30 biopsies from 17 patients with numerous types of gastrointestinal pathologies, including cancer, diverticulitis, and Crohn’s disease were imaged. In addition, using mouse models, we performed DµOCT imaging studies on tumors locally treated with chemotherapeutics delivered via custom implantable microdevices to observe the impact of those drugs on the tumor. Cyclical immunofluorescence staining was used to co-register ~20 markers on the same cross-sectional plane. We further demonstrate the utility of principal component analysis, K-means clustering, and convolutional long short-term memory (ConvLSTM) neural network for expanding the capabilities of DµOCT.

We introduce spectrally extended line field (SELF) technique to address some of limitations with the current OCTA, such as relatively small field of view, limited flow dynamic range, and motion artifacts. SELF-OCT acquires signals from multiple transverse positions simultaneously, providing an advantage in the transverse sampling rate. In addition, the allowable light exposure is higher so that there is more sensitivity budget for wide-field and functional imaging. We demonstrate that SELF technique enables improvement in field of view or transverse sampling density, flow dynamic range, sensitivity to slow flow, and motion tracking and correction in the human skin and retina in vivo.

We present the first dynamic full field optical coherence tomography (D-FFOCT) module which can be easily integrated into a commercial microscope. Benefitting from the optical standardization of commercial microscopy, we demonstrate three-dimensional live imaging experiments over time periods of minutes to hours to days thanks to the use of an incubator, able to maintain cell culture conditions. We show timelapse high resolution live images of retinal explants and organoids in disease modeling applications.

We demonstrate OCT-based intracellular motility imaging method, so-called dynamic-OCT (D-OCT), and its application for tumor spheroid-based drug testing. The volumetric tomography is captured in 52.4 s using our custom-designed scanning protocol, which repeatedly capture 32 frames at each location in the tissue. Two algorithms including logarithmic intensity variance (LIV) and late OCT correlation decay speed (OCDSl) were used for tissue dynamics visualization. The utility of our proposed method is investigated for the comparison of three types of anti-cancer drugs applied to human breast cancer (MCF-7) spheroids. The drug type dependent alterations of cell morphology and viability have been successfully visualized.

In traditional Chinese medicine (TCM), it is widely accepted that acupoints on meridian not only represent the treatment points but also serve as diagnostic points. We exploited our existing technology of multi-channel deep tissue microcirculation measurement known as diffuse speckle contrast analysis (DSCA) to see 1) whether a pressure stimulus on specific acupoint causes microcirculation change in other acupoints on the same meridian, and 2) whether the change is position-dependent. The experimental setup has been validated by cuff occlusion protocol, and then a pressure was applied on LI4 of human subjects several times with a regular time interval. We monitored, on acupoints of LI1, LI5, LI10, and ST25, not only the average blood flow index (BFI) but also the change of complexity in low frequency oscillation (LFO) signal which is known to be related to local tissue viability.

Interferometric near-infrared spectroscopy (iNIRS) noninvasively measures the optical and dynamical properties of the human brain in vivo. However, iNIRS uses single-mode fibers, which reduces the detected light throughput. Here, we demonstrate the parallel interferometric near-infrared spectroscopy (πNIRS) to overcome this limitation. In πNIRS we use multi-mode fibers for light collection and a high-speed, two-dimensional camera for light detection. With more than 8000 parallel channels, we can sense the cerebral blood flow and absorption changes with only 2-10 msec integration time (~100-500x faster than conventional iNIRS). This capability enabled us to monitor prefrontal cortex activation in humans in vivo.

Current report considers development of several Machine Learning (ML) classifiers for the quantitative functional imaging of human skin using polarized hyperspectral imaging. A validated optical model of has been combined with Monte Carlo-based computational approach and subsequently used in the training of Deep Learning methods. While demonstrating >98% accuracy for detecting important tissue features such as blood/oxygenation, pigment, etc. content we present detailed results of numerous data tests, training processes and the avenues for improving the performance of the developed techniques for newly captured data. The proposed techniques have a great potential to be implemented in low-coast clinical setting/wearable devices for e.g. monitoring and diagnosis of chronic skin ulcers and other relevant diseases.

In this work, we demonstrate a fiber-based interferometric Diffuse Optical Spectroscopy (iDOS) approach to obtain quasi-concurrent information at early and late times of flight via a simple fiber optic switch. Time-of-flight (TOF) filtering is enabled by reducing the effective temporal coherence of the laser source, here achieved through rapid wavelength tuning. Early and late TOFs are alternatively interrogated by optical switches that select between reference paths with short and long time delays, respectively. This approach is used on a human forearm to obtain quasi-concurrent deep and superficial blood flow index at baseline and during occlusion.

Live imaging of the mouse embryonic heart in 4D (3D+time) is critical for quantitative understanding of the early mammalian cardiac function and development. Optical coherence tomography (OCT) allows for 4D imaging of the cardiodynamics and hemodynamics in the mouse embryo with a unique and important spatiotemporal imaging scale. However, a limited accessibility of the reconstruction method to produce high-quality 4D OCT visualizations of the beating heart acts as a hurdle for a broader application of OCT in studying the embryonic heart. Here, we present an open-source, highly efficient, post-acquisition synchronization method to address this limitation.

We present three-dimensional, label-free renal tubular metabolism imaging by functional optical coherence tomography (OCT) including dynamics imaging method so-called “logarithmic intensity variance (LIV)” and OCT angiography (OCTA). Normal mouse kidneys and obstructed kidney models were investigated ex vivo. In the normal kidney, several vertical tails of high-LIV and hyper-OCTA signals were observed in the corresponding cross-sectional images. These signals formed pipe-like structures in the en face slab average projection images. In the obstructed kidneys, such anatomical pipe-like structures disappeared and instead, a circular shell at the edge of the renal cortex region was observed in the LIV.

At the early onset of a disease, the morphological changes manifest themselves on a cellular level (submicron scale). Label-free detection of nano-scale structural changes in sub-micron structure poses a significant challenge to researchers. A nano-sensitive OCT (nsOCT) technique permits to visualize 3D structure with nano-sensitivity to structural changes. Here, we present a brief theory and further development of the nsOCT approach, demonstrate different imaging modalities and unique abilities by monitoring internal nano-scale structural changes within different objects, including human skin in vivo. The presented results show new possibilities for study of structural changes, without the need for biomarkers or labels.

Imaging systems, including human vision, have a limited capability to separate spatial features, and these can also only be extracted over a limited depth range. The limits are related to the effect of diffraction and caused by the finite dimensions of the imaging optics and the geometry of the sensor. In my talk I will present novel photonic approaches to exceed the normal resolution and depth of focus limitations and show how those concepts can be applied in practical applications such as in microscopy, biomedical sensing, and ophthalmic devices to correct visual deficiencies.

Dynamic OCT angiography (OCTA) is an attractive approach for monitoring stimulus-evoked hemodynamics; however, a large dataset poses a great challenge to data processing. This study proposed a GPU-based real-time data processing pipeline for dynamic inverse SNR-decorrelation OCTA (ID-OCTA) with line-process rate of 133 kHz. Real-time processing enabled automatic optimization of angiogram quality, which improved the vessel SNR, contrast-to-noise ratio, and connectivity by 14.37, 14.08, and 9.76%, respectively. Furthermore, dynamic angiographic imaging of stimulus-evoked hemodynamics was achieved within a single trail in the mouse retina. Therefore, GPU ID-OCTA enables real-time and high-quality angiographic imaging and is particularly suitable for hemodynamic studies.