Abstract In vivo imaging challenges in cancer drug discovery include imaging liquid tumors and tracking immune cell activity after checkpoint immunotherapy for solid tumors. Current methods for real-time in vivo imaging of tumor models include bioluminescence imaging (BLI), which involves using genetically modified cell lines that can confound study data due to potential changes in the cell lines’ biology. This risk of potentially aberrant data is especially a concern with patient-derived xenograft (PDX) models. Our earlier work utilized targeted albumin-coated nanoparticles encapsulating rare earth-containing cores to image tumors in in vivo models of solid tumors and to image T cells in the tumor microenvironment. These biocompatible nanoparticles enable imaging without the need for genetic modification by emitting short-wave infrared (SWIR) light after excitation by a 980 nm light source. SWIR light travels through blood and tissue more efficiently than does visible bioluminescent light, providing deeper tissue illumination and sharper images. By functionalizing this technology with targeting antibodies, we were able to not only specifically image a range of solid tumors but also visualize CD8+ T cell activity around tumor sites. More recently, we have shown that our nanoparticles can be used to image and track an increase in liquid tumor burden within bone marrow and spleen over a three-week period in a lymphoma model. Briefly, U937 cells (50,000 per mouse) were injected i.v. into NSG mice on Day 0. Nanoparticles targeted to CD45 were injected on Days 10, 16, and 21, and the animals were imaged four hours post-injection using a SWIR-based imaging system. Tumor burdens were confirmed using flow cytometry. Since this initial lymphoma imaging study, we have improved our nanoparticle formulation by encapsulating the rare earth cores with a combination of biocompatible lipids and polymers, and by switching from a batch synthesis process to a microfluidics-based fabrication process. This microfluidics-based approach enables scalability and a high degree of batch-to-batch reproducibility. We tested these next-generation nanoparticles in vivo and showed that they are >5-fold brighter than our original albumin formulation, providing researchers with a highly sensitive imaging modality to interrogate the tumor microenvironment and to visualize liquid tumors in real time, all without the risks inherent in genetic modification of the tumors and immune cells being studied. Citation Format: Tiffany W. Leong, Pavel Abdulkin, Ameena A. Moghe, Vidya Ganapathy, Mark C. Pierce, Mark Ravera. In vivo short-wave infrared imaging for liquid tumor and immuno-oncology models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2137.

The shortwave infrared (SWIR) wavelength band (1,000-2,000 nm) has recently garnered increased interest due to its beneficial attributes for tissue imaging and spectroscopy. Lower scattering, increased sensitivity to endogenous water and lipid content, and the reduced influence of melanin offer new measurement opportunities and application areas. SWIR sensor technology is advancing at a rapid pace, fueling growth in academic and commercial instrument development. Here, we review recent progress in SWIR tissue imaging and spectroscopy, provide resources on tissue optical properties at SWIR wavelengths, and offer guidance on the utilization of models, methods, and instruments for SWIR measurements in tissue. One key finding is that while some efforts have exploited the lower scattering in the SWIR to achieve deep tissue imaging, this advantage is highly dependent on the specific choice of wavelength and the measurement geometry. Photon attenuation limits these advantages at wavelengths where water absorption is strong, especially in the context of higher-noise SWIR detectors. However, a growing number of clinical applications are emerging, especially those requiring in vivo water and lipid measurements, such as monitoring of tissue hydration, edema, lipid content, and others. There are further opportunities to expand this work toward more disease states while leveraging the low absorption of melanin in the SWIR. Since the field is currently limited by relatively few sources of tissue optical property data at SWIR wavelengths, we have compiled tabulated extinction values of SWIR chromophores from the literature. There is also a need for continued development of techniques for modeling photon propagation in tissue at SWIR wavelengths due to more moderate levels of tissue scattering compared to absorption, and methods to establish stable tissue-mimicking phantoms. The SWIR wavelength region is coming of age in the biophotonics community, with components and systems creating new opportunities in basic research and clinical applications once current challenges can be overcome.

IntroductionWater and lipid content in biological tissues are important biomarkers for understanding physiological processes and diseases. Spatial frequency domain imaging (SFDI) provides a non-invasive method to quantify these components over a wide field of view. This study introduces an LED-based shortwave infrared (SWIR) SFDI system to measure tissue hydration.MethodsThe system was first validated using water-lipid dilutions of known concentrations. Subsequently, SWIR-SFDI was applied to ex vivo porcine skin undergoing desiccation to observe the relationship between reduced scattering and measured water content changes. Finally, the dorsal hand was imaged in three human subjects before and after exercise to assess changes in tissue induced by perspiration.ResultsFor the water-lipid dilutions, the system accurately predicted chromophore concentrations, validating the approach. In the skin desiccation experiments, small decreases in water content led to pronounced reductions in the reduced scattering coefficient, whereas absorption showed limited sensitivity. In vivo results showed a marked decrease in reduced scattering following exercise, consistent with a loss of tissue hydration.DiscussionThe findings suggest that, under the specific circumstances tested here, the reduced scattering coefficient may be a more sensitive indicator of tissue hydration than absorption. This sensitivity to small changes in water content underscores the potential clinical utility of SWIR SFDI for non-invasive hydration assessment in biological tissues. This technique offers promising applications for clinical diagnostics and physiological monitoring.

Perspiration is a critical mechanism for regulating human core body temperature. In the development and assessment of antiperspirant products, objectively evaluating perspiration is essential for understanding sweat gland physiology and the factors determining sweat excretion onto the skin. This study introduces a novel approach using a FLIR A6750sc infrared thermal camera to capture real-time axillary sweat gland activity. The thermal images are enhanced using a contrast-limited adaptive histogram equalization algorithm, which maximizes thermal contrast and improves the visibility of active pore regions exhibiting slightly higher temperatures. To further analyze the data, a multilevel thresholding approach with four quantization levels is applied to segment and quantify individual pores and their activation areas in the projection map. By directly evaluating sweat pore activation in the axilla, this research investigates the impact of antiperspirant products on physiological responses. This non-invasive technique provides a more detailed and dynamic view of sweat gland function in the axilla compared to traditional methods, potentially offering new insights into the efficacy of antiperspirant products and the underlying mechanisms of sweat production.

Spectral imaging techniques have demonstrated the potential to quantify physiologically relevant tissue components by analyzing reflectance at multiple wavelengths. This study investigates the use of an RGBIR image sensor for estimating chromophore concentrations in skin and intraoral tissues. Raw pixel value measurements were calibrated to reflectance and then mapped to a diffusion model-based lookup table linking tissue reflectance to melanin, oxygen saturation, and water content. Initial benchtop experiments demonstrated changes in chromophore content consistent with an applied finger occlusion. The method was then translated to an intraoral imaging system, where reflectance changes were observed in response to erythema. These findings support the feasibility of RGBIR imaging for objective tissue assessment, with future work focusing on model refinement and clinical validation.

We present a novel 3D intraoral scanning approach using the TRIOS 4; 3Shape device, which enhances digital color correction, plaque detection, and stain identification. Aim: This study aims to improve diagnostic accuracy in 3D oral imaging by developing an objective, reproducible, and non-invasive methodology that addresses the limitations of subjective visual assessments and disclosing agents. Methods: Enhanced color accuracy was achieved by applying a calibrated color correction matrix to 25 3D scans of the Color Gauge Micro Target, ensuring improved color fidelity. Plaque detection was refined using clustering techniques to identify plaque regions, validated against benchmarks from plaque disclosing tablets. A stain detection algorithm was developed to provide a standardized assessment of surface discolorations. Results: The optimized color correction significantly enhanced the reliability of digital imaging, while the plaque detection algorithm effectively matched regions identified by disclosing agents. The stain detection method provided consistent classification in accordance with clinical labels. Conclusions: These advancements-enhanced color correction, plaque detection, and stain identification-significantly elevate the diagnostic potential of 3D intraoral imaging. By leveraging digital imaging techniques, this approach provides an objective and non-invasive solution that improves clinical accuracy and sets a new standard in oral healthcare.

Microscale systems have been underexplored in contemporary regenerative therapies developed to treat vision loss. The pairing of in vitro cell systems with optical fluorescent imaging provides unique opportunities to examine the infiltration of donor stem cells needed for successful transplantation therapies. A parallel eye device was developed to provide electric field (EF) stimulation to guide the migration of cells within 3D eye facsimiles synthesized from different ocular biomaterials. Cell infiltration within facsimiles was rapidly resolved using confocal microscopy to eliminate dependence on the cryostat sectioning commonly used for cell study. Moreover, EF stimulated galvanotaxis of donor cells within different depths of eye facsimiles. Optical imaging provided rapid resolution of z-stack images at physiologically appropriate depths below 500 microns. This study demonstrates that paired microscale–optical systems can be developed to elucidate understudied transplantation processes and improve future outcomes in patients.

In humans, perspiration regulates core body temperature. Therefore, objectively evaluating it is essential for studying sweat gland function and mechanisms, particularly in antiperspirant efficacy studies. Various approaches have been developed for measuring human perspiration and evaluating antiperspirant efficacy, but are unsuitable for robust and routine clinical testing applications. This paper shows how infrared thermography, utilizing both high- and low-resolution modes, functions as a multiscale imaging modality. The high-resolution mode extracts physiological parameters (respiratory similar to 0.3 Hz and heart rate similar to 1.0 Hz) and visualizes the reduction of the sweat pore radii (from 359 +/- 155 mu m to 161 +/- 47 mu m) after antiperspirant application, consistent with known mechanisms of pore plugging and constriction induced by aluminum salts. The low-resolution mode quantitatively maps sweat retention in underarm clothing. All study participants in a clinical trial showed reduced sweat retention on their T-shirts due to antiperspirants, with reductions ranging from approximately 37-97% and an average reduction of 77.7 +/- 22.1% using the developed methodology and tested antiperspirant. Overall, this non-invasive technique presents significant potential for clinical and personal care product evaluations, particularly in the early stages of product development.

Metastatic breast cancer remains a significant source of mortality amongst breast cancer patients and is generally considered incurable in part due to the difficulty in detection of early micro-metastases. The pre-metastatic niche (PMN) is a tissue microenvironment that has undergone changes to support the colonization and growth of circulating tumor cells, a key component of which is the myeloid-derived suppressor cell (MDSC). Therefore, the MDSC has been identified as a potential biomarker for PMN formation, the detection of which would enable clinicians to proactively treat metastases. However, there is currently no technology capable of the in situ detection of MDSCs available in the clinic. Here, we propose the use of shortwave infrared-emitting nanoprobes for the tracking of MDSCs and identification of the PMN. Our rare-earth albumin nanocomposites (ReANCs) are engineered to bind the Gr-1 surface marker of murine MDSCs. When delivered intravenously in murine models of breast cancer with high rates of metastasis, the targeted ReANCs demonstrated an increase in localization to the lungs in comparison to control ReANCs. However, no difference was seen in the model with slower rates of metastasis. This highlights the potential utility of MDSC-targeted nanoprobes to assess PMN development and prognosticate disease progression.

The development of a deep learning framework specifically designed for the analysis of intraoral soft and hard tissue conditions is presented in this paper, with a focus on remote healthcare and intraoral diagnostic applications. The framework Faster R-CNN ResNet-50 FPN was trained on a dataset comprising 4,173 anonymized images of teeth obtained from buccal, lingual, and occlusal surfaces of 7 subjects. Ground truth annotations were generated through manual labeling, encompassing tooth number and tooth segmentation. The deep learning framework was built using platforms and APIs within Amazon Web Services (AWS), including SageMaker, S3, and EC2. It leveraged their GPU systems to train and deploy the models. The framework demonstrated high accuracy in tooth identification and segmentation, achieving an accuracy exceeding 60% for tooth numbering. Another framework for detecting teeth shades was trained using 25,519 RGB and 25,519 LAB values from VITA Classical shades. It used a basic neural network leading to 85 % validation accuracy. By leveraging the power of Faster R-CNN and the scalability of AWS, the framework provides a robust solution for real-time analysis of intraoral images, facilitating timely detection and monitoring of oral health issues. The initial results provide accurate identification of tooth numbering and valuable insights into tooth shades. The results achieved by the deep learning framework demonstrates its potential as a tool for analyzing intraoral soft and hard tissue parameters such as tooth staining. It presents an opportunity to enhance accuracy and efficiency in connected health and intraoral diagnostics applications, ultimately advancing the field of oral health assessment.

Tooth color is an important parameter in cosmetic dentistry, to measure staining, effects of whitening products, or for matching the appearance of implants to neighboring teeth. The apparent color of teeth is affected by surface (extrinsic) and sub-surface (intrinsic) factors and is still assessed qualitatively by the dentist's visual impression. This study used a new color polarization camera to quantify tooth color. Recent commercial availability of snapshot color polarization cameras offers a new approach to rapidly quantify tissue color with depth selectivity. We applied this technology to quantify tooth color and are currently investigating its use in assessment of enamel demineralization.

Significance:Hyperspectral cameras capture spectral information at each pixel in an image. Acquired spectra can be analyzed to estimate quantities of absorbing and scattering components, but the use of traditional fitting algorithms over megapixel images can be computationally intensive. Deep learning algorithms can be trained to rapidly analyze spectral data and can potentially process hyperspectral camera data in real time. Aim:A hyperspectral camera was used to capture 1216 × 1936   pixel wide-field reflectance images of in vivo human tissue at 205 wavelength bands from 420 to 830 nm. Approach:The optical properties of oxyhemoglobin, deoxyhemoglobin, melanin, and scattering were used with multi-layer Monte Carlo models to generate simulated diffuse reflectance spectra for 24,000 random combinations of physiologically relevant tissue components. These spectra were then used to train an artificial neural network (ANN) to predict tissue component concentrations from an input reflectance spectrum. Results:The ANN achieved low root mean square errors in a test set of 6000 independent simulated diffuse reflectance spectra while calculating concentration values more than 4000× faster than a conventional iterative least squares approach. Conclusions:In vivo finger occlusion and gingival abrasion studies demonstrate the ability of this approach to rapidly generate high-resolution images of tissue component concentrations from a hyperspectral dataset acquired from human subjects.

In this study, we present an integrated stereoscopic and hyperspectral imaging system designed to overcome the limitations of traditional quantitative hyperspectral imaging, notably the dependency on precise camera-sample distance measurements. Our approach combines advanced depth-sensing technology with a compact hyperspectral camera, featuring integrated RGB sensors, to facilitate automated synchronization, system integration, and reconstruction through epipolar geometry and image co-registration. The system acquires hyperspectral data cubes along predefined camera trajectories, enabling full 3D hyperspectral representations via global alignment, a significant enhancement over conventional methods that lack depth resolution. This methodology has the potential to eliminate the need for strict camera-sample distance calibration and appends a morphological dimension to hyperspectral tissue analysis. The system's efficacy is demonstrated in vivo, focusing on non-contact human skin imaging. The integration of stereoscopic depth and hyperspectral data in our system marks a significant advancement in spectroscopic tissue analysis, with promising applications in telehealth, enhancing both the diagnostic capabilities and accessibility of advanced imaging technologies.

Hyperspectral imaging can capture light reflected from tissue with high spectral and spatial resolution. Fitting algorithms can be applied to the spectrum at each pixel to estimate tissue chromophore concentrations, including blood, melanin, water, and fat. Traditional fitting methods are computationally intensive and slow when applied over an entire image. This study developed an artificial neural network (ANN) to rapidly calculate tissue oxygenation, blood, and melanin content from hyperspectral images. Linearly polarized light from a halogen lamp was delivered through a ring illuminator placed 20 cm from the tissue surface. A 1024x1224 pixel hyperspectral camera captured diffusely reflected light through an orthogonal polarizer at 299 wavelengths between 400-1000nm. To train an ANN, diffusion theory was used to generate reflectance spectra from 440-800nm for a uniform tissue containing 24,000 random combinations of physiologically relevant concentrations of oxyhemoglobin, deoxyhemoglobin, melanin, and scattering. The ANN was then tested by generating another 6,000 reflectance spectra from diffusion theory using physiological values and comparing the chromophore concentrations output by the ANN to ground truth values. The ANN demonstrated a root-mean-square error less than 0.01 in predicting each chromophore concentration from reflectance spectra simulated by diffusion theory. An in vivo finger occlusion experiment demonstrated the ability of the system to quantify changes in oxygen saturation and blood volume. This work demonstrates a new deep learning approach to rapidly process hyperspectral image data and accurately quantify tissue components.

Abstract Although triple-negative breast cancer (TNBC) can be treated with anti-PD-1 checkpoint immunotherapy in combination with chemotherapy, there remains a challenge in effectively monitoring therapeutic responses. Current non-invasive clinical imaging tools to evaluate response to treatment are reliant upon measurements of tumor volume and may fail to distinguish true progression from increased immune cell infiltration. Invasive biopsy sampling and immunohistochemistry (IHC) can elucidate changes in the immune landscape of treated tumors, but these methods are not conducive to providing real-time information. This study presents shortwave infrared (SWIR) imaging as a potential tool to detect treatment-induced cytotoxic T lymphocyte (CTL) infiltration non-invasively and in real time using rare earth metal-doped nanoparticles encapsulated in human serum albumin nanocomposites (ReANCs). ReANCs were chemically conjugated with anti-CD8α antibodies as targeting ligands to facilitate binding of the nanoprobes to CTLs with high specificity in a syngeneic mouse model of breast cancer. After treating the mice with combination anti-PD-1 and doxorubicin, volumetric analysis of the mammary fat pad tumors did not show any significant impact of treatment compared to single treatment and untreated control mice. However, increased CTL infiltration in the tumors of mice that received combination treatment was detected by in vivo SWIR imaging. CTL infiltration was validated with ex vivo IHC staining, and a monotonic relationship was observed between SWIR fluorescence and CD8 positivity. IHC staining of other immune markers, including CD45, CD3, CD4, and PD-L1, showed that combination treatment may influence in the expression of these markers, presenting additional targets that could be imaged with ReANCs in the future. In conclusion, the increase in SWIR signal from CD8-targeted ReANCs in tumors treated with combination immunotherapy and chemotherapy and the relationship with IHC staining highlight the ability to use SWIR imaging for non-invasive assessment of changes in immune dynamics following treatment. Citation Format: Jay V. Shah, Jake N. Siebert, Xinyu Zhao, Shuqing He, Richard E. Riman, Mei Chee Tan, Mark C. Pierce, Edmund C. Lattime, Vidya Ganapathy, Prabhas V. Moghe. Non-invasive shortwave infrared imaging of cytotoxic T lymphocyte infiltration for monitoring responses to combination immunotherapy and chemotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4178.

To address an increasing demand for accessible and affordable tools for at-home oral health assessment, this paper presents the development of a low-cost intraoral camera integrated with a deep learning approach for image analysis. The camera captures and analyzes images of soft and hard oral tissues, enabling real-time feedback on potential tooth staining and empowering users to proactively manage their oral health. The system utilizes an Azdent intraoral USB camera with the Raspberry Pi 400 computer and Intel (R) Neural Computing Stick for real-time image acquisition and processing. A neural network was trained on a dataset comprising 102,062 CIELAB and RGB values from the VITA classical shade guide. Ground truth annotations were generated through manual labeling, encompassing tooth number and stain levels. The deep learning approach demonstrated high accuracy in tooth stain identification with a testing accuracy exceeding 0.6. This study demonstrates the capacity of low-cost camera hardware and deep learning algorithms to effectively categorize tooth stain levels with high accuracy. By bridging the gap between professional care and homebased oral health monitoring, the development of this low-cost platform holds promise in facilitating early detection and monitoring of oral health issues.

PDF file - 161K, Figure 1: Schematic diagrams of the imaging systems used in this study. Supplementary Figure 2: Multimodal imaging at the floor-of-mouth in a 73-year old female patient.