This PDF file contains the front matter associated with SPIE Proceedings Volume 11622, including the Title Page, Copyright information, and Table of Contents.

Machine learning offers a powerful set of tools to make widefield endoscopic imaging more quantitative. This presentation covers our work in estimating pixel size, topography, optical properties, and molecular chromophores using structured illumination and generative adversarial networks. We aim to create a computational endoscope that will improve computer-aided detection and diagnosis.

Deep Networks trained on one kind of data tend to perform poorly, on data that is beyond its training set. We believe this is because data sets tend to focus too directly on a specific task. We circumvent this by simulating various sinusoidal signal sums, with and without envelopes, along with blurred spike trains. We then add various noise to these signals during training to allow the networks to learn a denoising technique. Without using any real Raman or Brillouin data, our network successfully denoises and removes low frequency drifts from real experimentally acquired Raman and Brillouin data.

On-site pathology of the surface of resected tumors provides real-time assessments of surgical procedure and improve treatment outcomes. Tissue preparation is a major bottleneck of standard pathology procedures that can be overcome by virtual pathology imaging. Here, we use optical sectioning super-resolution structured illumination microscopy (OS-SR-SIM) with a low magnification objective. Low magnification objective provides a higher field of view, and super-resolution reconstruction retrieves lost resolution due to lower NA. We successfully resolved sub-resolution fluorescent beads in highly scattering media that were not resolvable with standard OS-SIM with an NA 0.45 objective. Human tissue testing is underway.

We propose a novel method for generating linear structured illumination using phase modulation. We implemented linear structured light using holographic technique. The computer generates double concentric annular slits with different radius, a prism phase is applied on the slits to tilts the beam that incidents on the slits away from the optical axis. Different annular beam produce Bessel beams with different axial wave vectors, linear structured light can be obtained from the interference between two Bessel beams. The period and phase of linear structured light can be adjusted by adjusting the radius and the initial phase difference of the double concentric annular beams. We theoretically and experimentally verify that the method can effectively generate linear structured light.

We present a novel approach, based on the use of an array of cameras with custom optics, which can capture snapshot stereoscopic gigapixel images across 1cm2 area at 1-micrometer half-pitch resolution. Our system uses a large space-bandwidth product objective lens to form an intermediate image, which is captured by 96 micro-cameras arranged in a flat array. Each camera records a 10-megapixel image from a unique section of the sample, which are then stitched to produce the final composite. Our system is well suited for applications in digital pathology and in vitro cell-cultures imaging.

Even though vascularization is recognized as an important factor in bone healing, there are limitations in terms of assessing vascularization in both preclinical and clinical scenarios. Diffuse optics have a potential to fill this gap by providing deep-tissue, non-invasive, longitudinal monitoring of total hemoglobin concentration, blood oxygen saturation and blood flow. Using rodent models with bone grafts or bone fractures, we will demonstrate the potential to build a prediction model for bone treatment efficacy assessment. In addition, the construction of a multi-modal diffuse optical imager to detect nonunion and monitor treatment in foot fractures will be introduced. Validation studies to compare diffuse optics with gold standard methods will be presented.

We used Diffuse Reflectance spectroscopy (DRS) to investigate functional changes within the tumor microenvironment when treated with different immunotherapy drugs. We injected the CT26 cell line into the flank of 35 mice that were assigned to 4 different groups: anti-PD-1, anti-CTLA-4, combination treatment, and IgG. The 4 groups were injected intraperitoneally on days 1, 4 and 7 (Day 1 – tumor at 80-120 mm3). DRS spectra were acquired simultaneously for 9 consecutive days and a lookup table (LUT)-based inverse model was implemented to decompose the spectral data. We found statistically significant differences in functional changes among the different immunotherapy groups.

We used diffuse reflectance spectroscopy to assess the effects of different radiation therapy doses on tumor microenvironment. We injected 4T1 cells into the flanks of 25 mice to develop tumor xenografts, and randomly distributed the animals into control and radiation groups, which received a dose of either 1, 2, or 4 Gy. Treatment started when tumors reached a volume of 200 mm3. DRS measurements were obtained prior to and 1 hour after radiation on the five days of treatment, and once a day thereafter. Our data demonstrates that DRS is sensitive to tumor microenvironmental changes after low doses of radiation.

Successful gastrointestinal surgery is based on the precise knowledge of the morphological, functional and metabolic state of the bowel wall at a specific time point. Current trends include the development of real-time, minimally invasive, label-free and rapid techniques for tissue assessment in combination with algorithms of data processing. The aim of the study was to evaluate the performance of trans-serous multimodal optical coherence tomography (MM OCT) and FLIM macro-imaging in detecting changes in microstructure, blood circulation and metabolism of intestinal wall caused by acute arterial ischemia in experiment. The study was supported by the Russian Science Foundation, project No. 19-75-10096.

In this study, we explore the use of red blood cell and hemoglobin autofluorescence ad potential long-term biomarkers for diabetes. We found that under 370 nm excitation, both red blood cells (RBC) and Hb fluorescence in the 420-600 nm region. At early time points following diabetic induction in rats, autofluorescence increases in lysed Hb is more dramatic compared to that of RBC. Moreover, we found significance variance of Hb autofluorescence despite relatively constant HbA1c levels. The results of our study suggest that with additional development, RBC and hemoglobin autofluorescence from may be used as long-term glycemic markers for monitoring diabetic complications in the clinical setting.

In this talk, we will discuss how scattered light can be used for noninvasive detection of invisible pre-cancer in organs such as the esophagus or pancreas which seem to have little in common. Nevertheless, since pre-cancer in many organs is characterized by certain common microscopic changes in the epithelial cells, such as the increase in nuclear size and nuclear density, we will show that light scattering signatures of those pre-cancers are quite similar, allowing for early cancer imaging and detection without the need for external markers. Light scattering signatures can also be used for sensing subnuclear and subcelluar structures, such as chromatin packing, organelle organization, and characterization of cell-derived exosomes. Nanoscale changes in the nuclear structure have been shown to play a critical role in genetic and transcriptional alterations and are a hallmark of neoplasia. We will discuss how the approach based on the combination of confocal microscopy and spectroscopy

Breast cancer is the second most common cancer among women in the United States. Heterogeneity in breast cancer treatment response across patients indicates that patient-specific treatment screens will reduce under-treatment of aggressive and recurrent cancers, while also sparing patients with non-aggressive disease toxicities due to overtreatment. Xenografts grown from patient-specific tissue in zebrafish present a novel platform for a medium to high-throughput anti-cancer drug screen to individualize patient therapy. The goal of this project is to develop an optical imaging anti-cancer drug screen to evaluate patient-specific zebrafish tumors. Trastuzumab (anti-HER2 antibody) responsive and resistant breast cancer cells were injected into 48 hour-post-fertilization zebrafish embryos. Tumors were established for 24 hours and then fish were treated with a panel of breast cancer drugs and drug combinations. Autofluorescence lifetime images with single-cell resolution of the fish tumors were acquired at 24, 48, and 72 hours of drug treatment. Wide-field mCherry fluorescence and bright-field images with fish-level resolution were acquired prior to treatment (t=0 hour) and at 72 hours of drug treatment. Individual tumor responses were determined from the whole-fish bright-field and mCherry fluorescence images. Substantial differences in autofluorescence lifetime features, including optical redox ratio and mean NADH lifetime, were found between drug responsive and resistant tumors. These results suggest autofluorescence lifetime imaging is predictive of anti-cancer drug response in zebrafish xenografts

Fluorescence lifetime imaging (FLIM) is a powerful tool to quantify local changes in cell metabolism without loss of spatial resolution. Here, we adopted FLIM to characterize spatio-temporal metabolic changes occurring during collective cell migration. Using an established in vitro system, we measured biomechanical and metabolic changes during migration of epithelial cell monolayers. We developed a custom image analysis pipeline that combines machine learning segmentation and curve fitting analysis to analyze FLIM data. Our findings – which were validated via separate measurements of cytoplasmic redox ratio, glucose uptake, and mitochondrial membrane potential – are consistent with a glycolytic shift during collective cell migration.

Vibrational spectroscopy is widely used for a large variety of sensing and imaging applications. Nonlinear optical interactions using ultrashort laser pulses can facilitate selective coherent excitation of molecular vibrational modes, such as stretching and bending C-H and O-H vibrational modes. However, such spectroscopic probes are often based on the assumption that the energy distribution is static throughout the system, neglecting complex molecular dynamics and the effects of local nano- and microenvironments. To tackle this challenge, we developed a coherent control technique with a special focus on the temporal evolution of molecular vibrational modes. By utilizing a high-dynamic range detection, we demonstrate molecular vibrational dynamics and the environmental effects with multidimensional spectroscopic sensing. Such capability promises a broad range of sensing and imaging applications in biology, materials, and chemical sciences.

The recognized need to develop better clinical approaches for detection of epithelial cancers and potentially malignant lesions than currently used has motivated work in development of noninvasive fluorescence imaging devices. While individual large area imaging and microscopic techniques are promising, recent trends have explored combinations that could merge strengths. The study will discuss a workflow to combine strengths of label-free nonlinear optical microscopy (NLOM) which has shown promise for optical biopsy but is limited in scannable area with widefield autofluorescence microscopy providing large surface area assessment, in studies conducted in both a hamster model for oral neoplasia and inflammation and in surgical oral cancer specimens.

Diffuse optical spectroscopy (DOS), also known as near-infrared spectroscopy (NIRS), offers sub-micromolar sensitivity to tissue composition, perfusion, and oxygen metabolism; low patient risk since DOS requires neither ionizing radiation nor contrast agents; and relatively low-cost instrumentation. Consequently, DOS methods are applied in virtually all major areas of clinical research including neurologic disease, cancer, cardiovascular disease, metabolic disease, and trauma/critical care. Compared to other medical imaging technologies, DOS allows for ultracompact integration that can enable handheld, wearable, and even implanted DOS-based sensing strategies. In this presentation, we discuss our work to create a handheld frequency-domain quantitative tissue imager and a tumor-implantable DOS-based sensor roughly the size of a standard breast radiological clip.

We present an update on our high optode-density continuous-wave (CW) wearable diffuse optical device for the investigation of hemodynamic responses of locally advanced breast tumors during neoadjuvant chemotherapy (NAC). The device consists of a rigid-flex substrate with 32 LEDs at two wavelengths and 16 detectors. Measurements during a cuff occlusion indicate that the probe can quantify hemodynamics temporally, and measurements on spatially-complex flow phantoms have validated the ability to reconstruct spatial contrast. A normal volunteer study is currently ongoing, and preliminary results (N=7 volunteers) indicate that paced breathing hemodynamics can be quantified in healthy subjects.

Human female breast is composed of skin, fibrous tissue, breast glands and fat. Breast cancer is a malignant tumor that occurs in the epithelial tissue of breast glands. The breast is not an important organ for maintaining human life. Breast cancer in situ is not fatal; however, because breast cancer cells lose the characteristics of normal cells, the connections between cells are loose and easy to fall off. Once the cancer cells fall off, the free cancer cells can spread throughout the body with the blood or lymph fluid, forming metastases, and endangering life. Breast cancer has become a common tumor threatening women's physical and mental health. Therefore, studying the interaction between laser and breast tissue and breast tumors has important theoretical and practical significance for the treatment of breast cancer. To this end, this research uses the commercial finite element simulation software COMSOL Multiphysics to develop a two-dimensional numerical simulation model based on finite element, which studies the propagation and heat transfer of light in the breast of breast cancer patients. In this study, the model consists of four parts: 1) water layer; 2) breast; 3) breast tumor; 4) short pulse laser source (wavelength is 840nm). The laser point source is located in the middle of the water layer above the breast tissue to irradiate the breast and tumor. Simulate the propagation of light in the breast and tumor by solving the diffusion equation. The temperature changes of breast tissue and breast tumors are obtained by solving the biological heat transfer equation. This research helps to understand the spread of light in human breasts and breast tumors and the interaction between the two, and has certain theoretical guiding significance for the research and treatment of breast cancer.

Light- and ultrasound-based technologies are growingly used for brain interrogation and modulation of neural activity at various spatial and temporal scales. We recently developed functional optoacoustic neuro-tomography (FONT) technique to enable the recording of multiple hemodynamic parameters at high spatial resolution across the entire mouse brain. The method was also applied for tracking rapid neural dynamics using genetically encoded calcium indicators, thus overcoming the longstanding penetration barrier of optical microscopy in scattering brains. FONT thus enables large-scale neural recording at penetration depths and spatio-temporal resolution scales not covered with the existing micro- and macro-scopic functional neuroimaging techniques. Hybrid implementations have further been developed to allow for real-time monitoring of the effects of transcranial ultrasound stimulation of the living brain.

Active line scan imaging (LSI) with spatial gating is explored for wide-field, sub-diffuse sensing of scattering microtextures. Spatial gating involves masking detector pixels far from the laser line of incidence on a turbid target. The active line scan provides broadband spatial frequency modulation, and the spatial gating effectively high pass filters the reflectance. The major benefits of LSI are that of high dynamic range signal preservation and high contrast-to-noise when imaging at high spatial frequencies. As such, LSI presents as an inherently high sensitivity and high dynamic range way to image microscopic scattering features on the surfaces of turbid media.

The core idea of the spectracoustic technique is the development and fabrication of a common probe for two modalities, the infrared spectroscopy and ultrasonic μTomography, for its application as a non-invasive analysis technique for tissue classification. The acquired and fused data from both modalities, provide a spectroscopic mapping tomographic image. The probe permits the excitation of the under-study object using both techniques simultaneously or in serial mode. Through the ultrasonic transducer of the probe, ultrasonic wave pulses are transmitted in the under-study tissue. Parallel to the path, used for the excitation of the piezoelectric transducer, a fiber optic bundle path is also designed in order to illuminate the under-study tissue. The reflected waves are transmitted back through the fiber optic bundle. The path used for emitting both the ultrasonics and infrared waves is filled with a special gel material in order for the ultrasonic probe to be coupled with the tissue. The infrared spectrum of this material is used as background spectrum form the infrared modality in order to be corrected from the acquired spectrum. By scanning a tissue in a specific region of interest, the incrementation of tomographic and spectroscopic data is achieved.

We developed a Bessel-beam photoacoustic microscopical simulation platform by using the k-Wave: MATLAB toolbox. The simulation platform uses the ring slit method to generate Bessel beam. By controlling the inner and outer radius of the ring slit, the depth-of-field (DoF) of Bessel beam can be controlled. And the large volumetric image is obtained by point scanning. The simulation experiments on blood vessels was carried out to demonstrate the feasibility of the simulation platform. This simulation work can be used as an auxiliary tool for the research of Bessel-beam photoacoustic microscopy.

The dynamic breast skin temperature evaluation is investigated in the present work for the diagnosis of breast cancer at an initial stage. In particular, a realistic breast model, including a relatively small-sized tumor, is designed and thoroughly analyzed numerically in terms of the biological heat transfer equation. Initially, the skin surface temperature is measured for a steady-state scenario at normal conditions, indicating that a tumor behind the papilla corresponds to an almost physiological thermogram. For this reason, appropriate cooling stress is applied to capture the time-response of skin temperature throughout this procedure. Numerical results highlight that the contrast is enhanced thus facilitating an early diagnosis of breast cancer.

Photoacoustic imaging is a promising technique that combines optical contrast with ultrasonic detection to map the distribution of the absorbing pigments in biological tissues. It has been widely used in biological researches, such as structural imaging of vasculature, brain structural and functional imaging, and tumor detection. Photoacoustic microscopy (PAM) is an important branch of Photoacoustic imaging that can achieve spatial from the micron to submicron level. However, Common photoacoustic microscopy can only be fixed on a single scale (single resolution) for imaging, which makes it difficult to obtain rich information of biological tissue, and cannot adapt to different imaging needs. Here, we developed multi-scale photoacoustic microscopy by integrating optical-resolution photoacoustic microscopy and the acoustic-resolution photoacoustic microscopy using multi-dimension optical fiber. This method mainly uses a single mode fiber and a multimode fiber with same length to split a single laser pulse into two sub-pulses. Finally, the two laser pulses are focused on the sample by the objective lens, which can generate a focal spot covering a few microns to tens of microns near the focal plane. The size of the focal spot is determined by the core diameter of the fiber, and the different size of the spot leads to different lateral resolution, Therefore, the system can achieved both optical resolution (4 μm) and acoustic resolution (40 μm). The multi-scale imaging performance of the system is verified by imaging the blood vessels of mouse and cerebral vascular. The multiscale imaging capability of our system may fulfill different requirements in biomedical researches and facilitate multiscale interpretation in system biology.

Photoacoustic imaging is a new non-destructive medical imaging technology based on photoacoustic effect. It can reflect the difference of light absorption energy by detecting photoacoustic signal. At present, the analysis methods of photoacoustic signals in biological tissues can be divided into two categories, namely, time-domain analysis of signals and frequency-domain analysis of signals. In time domain analysis, the envelope of the received photoacoustic signal is usually used to reconstruct the image. However, due to the influence of various external factors, the time domain signal cannot accurately reflect the characteristics of the absorber itself. Here, photoacoustic spectrum analysis was performed by using k-Wave to obtains the relationship between the structure, size, density of the absorber and the photoacoustic spectrum. Firstly, the relationship between the size of absorber and the photoacoustic spectrum is studied, and the slope and intercept are used to analyze the spectrum. Conversely, the relationship was used to predict the size of the absorber Finally, we used this relationship to predict the size of blood vessels.