Significance:Maximizing brain tumor resection while preserving neurological function remains a neurosurgical challenge. Current intraoperative tools, including MRI, ultrasound, and frozen section, disrupt workflow and provide limited real-time molecular information. Aim:This review synthesizes evidence for Raman spectroscopy as an intraoperative tool in glioma and brain metastasis surgery, comparing it against established alternatives and framing findings within a diagnostic hierarchy from analytic validity through clinical utility to outcome benefit. Approach:Evidence is arranged by surgical setting: in vivo handheld fiber-optic probes and ex vivo platforms including Raman microscopy, stimulated Raman scattering (SRS) imaging, and visible resonance Raman. Results:In vivo multicenter studies report > 85 % per-measurement sensitivity and specificity for tumor versus normal classification, with nondestructive analysis in ∼ 2 s . Molecular marker detection, including IDH mutation status, has been demonstrated predominantly ex vivo. Ex vivo SRS generates histology-quality images and can classify selected molecular alterations within ∼ 90 s , though this molecular capability is not clinically validated. Conclusions:While multicenter studies support the analytic validity of Raman-based tissue detection, clinical adoption requires prospective outcome trials, standardized acquisition protocols, and regulatory advancement.
SignificanceAccurate identification of residual tumor during head and neck cancer surgery is essential for preventing local recurrences and improving patient outcomes, yet current methods for intraoperative margin assessment remain limited. Fluorescence lifetime imaging (FLIm), a label-free optical technique, has shown strong potential for distinguishing cancerous from noncancerous tissue in vivo. However, the effect of tissue resection on autofluorescence properties is not well defined, raising the question of whether diagnostic signatures establishedin vivo translate reliably to ex vivo specimens.AimThe aim is to characterize how fluorescence lifetime properties change between in vivo and ex vivo conditions across multiple head and neck anatomical sites and to assess how these changes affect diagnostic separability between cancerous and noncancerous tissue.ApproachA custom fiber-based, point-scanning FLIm system with phasor-based analysis was used to characterize fluorescence lifetime features in vivo and ex vivo across oral tongue, base of the tongue, and palatine tonsil tissue from 75 patients undergoing oropharyngeal surgery. A separability index based on phasor distributions was used to quantify cancer versus noncancer contrast across multiple spectral channels.ResultsResection-induced changes in decay dynamics and fluorescence lifetime values varied by anatomy: oral tongue and tonsil showed reduced contrast ex vivo, whereas the base of tongue demonstrated improved separation after excision. The base of the tongue showed a higher separability index ex vivo (0.43 versus 0.15 in vivo), whereas oral tongue and tonsil maintained superior contrast in vivo (0.27 versus 0.18 and 0.46 versus 0.28, respectively).ConclusionsSurgical resection substantially alters autofluorescence signatures in an anatomy-dependent manner, emphasizing the need to train and validate diagnostic algorithms independently for in vivo surgical guidance and ex vivo pathology assessment.
Significance:Estimation of tissue optical properties in workflow- and resource-constrained clinical settings is required to support photodynamic therapy (PDT) treatment planning, yet large-scale clinical translation remains limited by spectroscopy probe cost, complexity, and the need for probe-specific inverse models. Single-fiber reflectance (SFR) spectroscopy offers a simple and low-cost option, but previous implementations relied on customized angle-polished fibers. Aim:The objective was to develop a proof-of-concept SFR spectroscopy system employing an unmodified, United States Food and Drug Administration (FDA)-approved clinical optical fiber for quantitative optical property extraction. Approach:A semi-empirical photon pathlength model was used to recover the absorption spectra μ a ( λ ) and reduced scattering spectra μ s ' ( λ ) from tissue-mimicking phantoms containing methylene blue (MB) as the absorber and Intralipid-20% as the scatterer. Results:Optical property retrieval from measurements acquired using three nominally identical fibers demonstrated clinically sufficient accuracy across fibers without requiring model adaptation. Across all fibers, MB concentration ( C MB ) was recovered with a root mean square error (RMSE) of 0.6 to 0.7 μ M ( p = 0.88 among fibers), while μ s ' ( 665 nm ) was recovered with an RMSE of 1.5 to 2.2 cm - 1 ( p = 0.68 among fibers). Conclusions:We demonstrate the potential of SFR spectroscopy using a standard clinical optical fiber as a scalable and clinically useful tool for tissue optical property estimation to support PDT treatment planning.
Significance:The Monte Carlo method for light transport is widely accepted as an accurate method for simulating light propagation in a scattering medium. Its use in optical tomography, however, suffers from inherent stochastic noise. This noise is present in both evaluations of the forward model, as well as in the search direction of the minimization algorithm used for image reconstruction. Aim:We aim to utilize machine learning to compensate for the stochastic Monte Carlo noise in the reconstruction of absorption and scattering in optical tomography. Approach:An iterative image reconstruction algorithm is proposed. The algorithm uses convolutional neural networks in a stochastic Gauss-Newton update when estimating absorption and scattering coefficients. Results:The methodology is evaluated using numerical simulations and compared against the conventional stochastic Gauss-Newton algorithm in optical tomography. It is demonstrated that the methodology can be used to compensate for image reconstruction artifacts caused by the stochastic noise. Conclusions:The proposed machine learning approach can be used to compensate for stochastic noise in Gauss-Newton iterations, and it enables reconstruction of absorption and scattering with a significantly lower number of photons than a conventional stochastic Gauss-Newton algorithm.
Significance:Cardiac panoramic optical mapping is a powerful approach for studying action potential dispersion and mapping arrhythmia triggers and propagation pathways over the entire surface of the heart. However, tissue type (muscle, connective tissue, and infarct scar) is also important for interpreting mapping data and is difficult to identify using optical mapping data alone. Aim:Panoramically map transmembrane potential and tissue type from the surface of infarcted hearts for correlative analysis of cardiac structure and function. Approach:We developed a multimodal panoramic imaging system to map epicardial tissue type (determined by collagen content) using a line-scan hyperspectral camera and a precision stage to translate and rotate the heart while illuminating the epicardial surface with UV light. Transmembrane potential was subsequently optically mapped by imaging a potentiometric probe with four high speed CMOS cameras position around the heart. The epicardial surface was reconstructed for each heart using images acquired every 3.6 deg of rotation, onto which hyperspectral and optical mapping data were texture mapped. All cameras were registered to one coordinate frame using a calibration procedure. Results:This system combines, for the first time, high-resolution hyperspectral imaging with optical mapping for quantitative correlative tissue structure-function analyses. It was used to study excitation wave propagation and action potentials across the surface of perfused rat hearts having a four-week-old infarct. The spectral band of collagen fluorescence (400 to 520 nm) revealed infarcted and border zone tissue. PVCs and reentrant activity were observed in 3 of 4 hearts at S1-S2 pacing intervals between 80 and 65 msec (S1 = 150 msec). PVCs originated near the infarct border and propagated around the infarct. Using the integral of spectral intensity from 400 to 435 nm, a k-means clustering algorithm classified each mapped site as either healthy, border zone, or infarcted tissue. Average action potential duration within those tissue types was longest for infarcted tissue, shorter for border zone tissue, and shortest for healthy tissue, a preliminary result that is consistent with the effect of an infarct on ventricular electrophysiology. Conclusions:This work demonstrates that panoramic hyperspectral mapping of tissue type and transmembrane potential is a powerful approach that enables functional mapping data to be analyzed within the context of local tissue type (healthy, infarct, and border) in living hearts.
Significance:Optical coherence tomography (OCT) is widely used for the diagnosis of retinal diseases. However, deep learning models trained on a single dataset often degrade when deployed across scanners and clinical sites due to device-dependent speckle variability and acquisition differences, limiting their reliability in real-world screening. Aim:We aim to develop a lightweight deep learning framework that leverages speckle characteristics in OCT images to improve cross-scanner generalizability for retinal disease classification while preserving real-time inference efficiency. Approach:We propose NA-DyCNN, a noise-aware dynamic convolutional neural network that minimizes the expected classification risk over multiple stochastic realizations of multiplicative speckle perturbations and regularizes the dynamic routing mechanism to produce scanner-invariant kernel mixtures. The framework was evaluated using over 105,000 B-scans from three heterogeneous OCT cohorts under strict zero-shot cross-dataset transfer. Results:NA-DyCNN consistently outperformed lightweight baselines across four zero-shot cross-dataset transfer scenarios, achieving up to 92.87% accuracy, a weighted F 2 score of 92.89%, and Cohen's κ of up to 0.896, demonstrating improved robustness and generalization under cross-dataset shifts. The model maintained high efficiency with only 0.4 M parameters and an inference latency of 0.53 ms per B-scan on an NVIDIA GB10 GPU. Conclusions:Modeling postacquisition speckle variability during training improves the generalizability of OCT classifiers without increasing the inference cost, thereby enabling the more reliable deployment of artificial intelligence (AI)-assisted retinal screening across heterogeneous imaging systems.
Significance:Maximizing safe tumor resection remains a major challenge in brain tumor surgery due to the lack of reliable real-time intraoperative contrast between tumor and healthy brain tissue. Label-free optical methods based on endogenous tissue fluorescence could provide simple, noninvasive feedback to guide resection while avoiding the logistical and regulatory limitations of exogenous fluorophores. Aim:The study aims to investigate whether endogenous near-infrared (NIR) fluorescence can differentiate tumor from normal brain tissue in vivo and to develop a wide-field fluorescence and reflectance imaging system for label-free visualization of brain tumor specimens ex vivo. Approach:We analyzed in vivo data acquired with a hand-held near-infrared spectroscopy probe from 26 patients (430 measurements) to quantify endogenous fluorescence differences between glioblastoma or astrocytoma and normal brain. Based on these findings, we built a wide-field imaging prototype using 808-nm laser excitation, red-shifted fluorescence detection with an InGaAs short-wave infrared camera, and co-registered reflectance imaging for ratiometric normalization. The system was validated with ex vivo human surgical specimens (14 samples from five patients). Results:Endogenous fluorescence intensity measured in vivo was significantly higher in normal brain tissue compared with tumor, yielding positive predictive values of 85 to 88% for tumor and a negative predictive value of 96% for normal tissue. Ex vivo wide-field imaging reproduced this contrast: normalized fluorescence was lower in tumor regions than in histologically confirmed normal areas, consistent with metabolic differences observed in vivo. Conclusions:Endogenous NIR fluorescence provides intrinsic contrast between normal and tumoral brain tissue. The demonstrated corrected wide-field fluorescence imaging approach could offer a feasible, label-free method for real-time visualization of tumor margins, supporting its future clinical translation as an intraoperative guidance tool.
Significance:Reversibly switchable optoacoustic proteins (rsOAPs) are a promising candidate for sensitive and quantitative optoacoustic (OA) imaging of genetically modified cell populations, such as chimeric antigen receptor (CAR) T-cells used in cancer immunotherapy. Although detection of rsOAPs has been demonstrated using classical machine learning approaches, there is still a need for higher detection sensitivity and more accurate quantification. Aim:We aim to develop deep learning approaches for improving the sensitivity of the detection and accuracy of quantification of rsOAPs with OA imaging and a 3D simulation framework to create synthetic datasets for machine learning experiments. Approach:We developed a forward model to generate labeled synthetic OA images of rsOAPs that takes into account light transport, acoustic wave propagation, and light-driven transitions between two different forms of the proteins. We used the synthetic images to train and evaluate machine learning models, including two convolutional neural networks and a transformer neural network, on the binary semantic segmentation and pixel-level prediction (regression) of the spatial distribution of the rsOAPs. Results:With this dataset, fine-tuned convolutional and transformer neural networks substantially outperformed classical machine learning approaches in the binary semantic segmentation of rsOAPs within the inhomogeneities (regions representing tumors), increasing the sensitivity from around 0.38 to 0.55 for noiseless data and from around 0.15 to 0.53 under a high level of stochastic noise ( SNR dB = 8.8 ) while maintaining a specificity of around 0.84, down to protein concentrations of order 10 - 8 M . Conclusions:A new methodology for the generation of synthetic OA imaging data of rsOAPs is presented, and the feasibility of deep learning for the accurate semantic segmentation and quantification of rsOAPs in OA imaging is demonstrated.
Significance:Optical coherence elastography (OCE) is an emerging biomedical imaging technique for mapping the micro-scale mechanical properties of tissue. Phantoms are vital for assessing OCE imaging performance in a controlled and systematic manner. Although the use of phantoms in OCE has been widely demonstrated, there is no consensus on suitable phantom materials or fabrication methods, limiting reproducibility and inter-laboratory comparison of OCE techniques. Aim:Our aim is to establish a unified framework for selecting, fabricating, and characterizing OCE phantoms by systematically evaluating the mechanical, optical, and structural properties of silicone, agar, and gelatin, the three most widely used OCE phantom materials. Approach:We conducted a literature review of phantom fabrication methods reported in OCE studies published between 1998 and 2025, comprising 223 papers, and identified silicone, agar, and gelatin as the most widely used OCE phantom materials. We experimentally characterized the elasticity, viscoelasticity, and attenuation coefficient of homogeneous phantoms fabricated from each material and investigated the effects of shelf life and optical scatterer concentration on mechanical properties. We further fabricated inclusion phantoms and tissue-mimicking surface roughness phantoms derived from optical coherence tomography (OCT) scans of human breast tissue using all three materials and evaluated their imaging performance using a compression OCE technique, quantitative micro-elastography (QME). Results:The elastic (tangent) modulus ranged from 7.2 to 175.9 kPa for silicone, 9.7 to 229.1 kPa for agar, and 3.5 to 29.4 kPa for gelatin, while OCT attenuation coefficients ranged from 0.2 to 13.1 mm - 1 across the three materials. Stress relaxation testing revealed distinct viscoelastic signatures, with relaxation time constants ranging from ∼ 2 s in silicone to over 1600 s in gelatin. We demonstrated that optical attenuation in these materials can be varied independently of mechanical properties. Inclusion and surface roughness phantoms were successfully fabricated from all three materials, with QME measurements revealing material-specific fabrication artifacts and demonstrating that surface topography can produce erroneous mechanical contrast in mechanically uniform materials. Conclusions:This study provides a framework to guide phantom material selection, fabrication, and characterization in OCE, identifying silicone as best suited for durable, predominantly elastic phantoms; agar for viscoelastic phantoms across a broad elasticity range; and gelatin for soft, highly viscoelastic phantoms. We believe that the results presented here will support standardization in OCE phantom development and will enable more detailed analysis, validation, and comparison of OCE techniques.
Significance:Maximum permissible exposure limits defined in ANSI Z136.1 are broadly protective, but they do not explicitly account for pigmentation differences in skin. As a result, the lack of experimental data quantifying the effect of pigmentation on injury thresholds constrains accurate interpretation when melanin is a primary optical absorber, which has direct implications for the performance of optics-based imaging methods. Aim:We experimentally determined the pigmentation-dependent laser exposure dose that produces a visible injury in 50% of exposed sites (i.e., ED 50 injury thresholds) with nanosecond, 750-nm-wavelength pulses, and we provide the first known quantification of relationships between melanin index (MI) and individual typology angle (ITA) or erythema index (EI) through colorimetric measurements. Approach:Fifty total abdominal skin sites per swine were identified on three swine with different pigmentation phenotypes (median ITA values per swine ranged - 18 ° to 87°). Single-laser pulses with 5-ns pulse duration, 750-nm-wavelength, and energies of 35, 40, 50, 60, and 65 mJ per pulse were delivered to 10 sites per energy level per swine. Laser-exposed sites were visually assessed at early post-exposure intervals, followed by colorimeter measurements of MI, ITA, and EI. Probit regression analysis was used to compute the ED 50 thresholds, followed by a polynomial regression analysis conducted to relate ITA to MI and EI to MI. Results:Dark-skinned swine ( median ITA = - 18 ° ) exhibited a 27.7% lower ED 50 ( 323.94 mJ / cm 2 ) than light-skinned swine ( median ITA = 69.5 ° , ED 50 = 448.36 mJ / cm 2 ), which is consistent with increased melanin-mediated absorption in the darker-skinned swine. The lightest-skin swine ( median ITA = 87 ° ) did not develop visible erythema within the tested exposure range, preventing ED 50 estimation. Mathematical MI-ITA and MI-EI relationships revealed pigmentation-dependent optical and inflammatory responses across the three swine. Conclusions:The measured differences in ED 50 values quantify the impact of skin pigmentation on laser-induced injury risk and demonstrate that darker skin has greater susceptibility to damage with the investigated laser parameters, relative to lighter skin. This foundational work addresses a previously uncharacterized source of biological variability that is relevant for the interpretation of laser safety margins and the design and evaluation of optics-based systems.
Significance:Precise determination of corneal refractive indices is essential for accurate refractive power calculations, personalized refractive surgery planning, and glaucoma management. However, current clinical methods fail to distinguish between the group refractive index ( n g ) for thickness quantification and the phase refractive index ( n p ) for optical power calculations, often relying on population-averaged constants that mask individual heterogeneity. Aim:We aim to develop a multimodal measurement system integrating spectral-domain optical coherence tomography (SD-OCT) and confocal scanning for the precise extraction of the n g , n p , and dispersion coefficients of corneal tissues. Approach:A dual-modality platform sharing an 860 nm source was constructed to jointly capture the optical path length (OPL) and confocal distance. A two-step progressive workflow was implemented: first, the n g and thickness were independently determined via the OPL method to serve as target ground truths for extracting tissue-specific dispersion through iterative optimization; second, these quantified priors were utilized to jointly decouple the in situ n g , n p , and thickness of intact corneas. Results:System validation using optical window plates yielded a relative error of only 0.01% to 0.09% for n g and 0.17% to 5.74% for dispersion coefficients. For biological tissues, we report the first direct experimental measurement of phase dispersion for human lenticules ( - 0.014468 ± 0.004060 μ m - 1 ) and porcine sections ( - 0.010223 ± 0.002811 μ m - 1 ), where the human data showed high consistency with the theoretical Cauchy model ( - 0.014186 μ m - 1 ). Utilizing these priors, the system successfully decoupled the in situ parameters of whole porcine eyes, yielding n g of 1.3845 ± 0.0006 and an n p of 1.3757 ± 0.0006 . Conclusions:This technique successfully resolves the n g and n p decoupling challenge in corneal tissues, overcoming the limitations of relying on a single equivalent refractive index. By utilizing an empirically measured corneal dispersion coefficient as a mathematical constraint, the system enables the nondestructive decoupling of physical thickness, as well as phase and group refractive indices in situ. This breakthrough provides essential physical parameters to enhance clinical pachymetry precision and optimize personalized refractive surgery protocols.
Significance:Digital twins are transitioning from conceptual models to operational frameworks that link measurement, prediction, and intervention in biomedicine. However, most biomedical digital twin efforts remain fragmented, with limited integration across biological scales, sensing modalities, and clinical decision points. Biophotonics provides a uniquely suited measurement foundation for biomedical digital twins by enabling quantitative, physics-grounded, and longitudinal noninvasive measurements spanning molecular, cellular, tissue, organ, and whole-body scales. These capabilities position photonics as a foundational measurement layer for next-generation biomedical digital twins. Aim:To synthesize current advances and future opportunities in optical digital twins and to establish a unifying framework for how photonic sensing can support digital twin architectures for disease diagnosis, therapy guidance, prevention, continuous monitoring, and interventional healthcare. Approach:This white paper summarizes perspectives presented at the annual meeting of the international society for optics and photonics (SPIE Photonics West), in the session "Digital Twins as New Approach Methodologies (NAMs) in Biophotonics." We review five complementary implementations of the digital twin paradigm: (i) virtual tissue staining for histopathology, which combines label-free optical imaging with machine learning-based inference to generate clinically interpretable representations with uncertainty quantification and validation; (ii) cell-level metabolic digital twins that use autofluorescence and redox imaging to predict patient-specific therapeutic responses in tumor organoids and immune cells under controlled perturbations; (iii) therapeutic digital twin frameworks for radiation therapy, in which Cherenkov imaging and radiation chemistry sensing verify treatment delivery and enable biophysical model correction and personalization; (iv) personalized optical digital twins for continuous monitoring that integrate longitudinal photonic sensing with physiological and contextual data to support early detection, prevention, and adaptive care; and (v) personalized digital twins for interventional healthcare. Results:Across these diverse applications, a common digital twin architecture emerges. Optical measurements define patient state, inference models translate measurements into predictions, therapeutic interventions perturb the system, verification measurements constrain and validate execution, and longitudinal sensing continuously updates the twin over time. The reviewed examples demonstrate that optical measurements can serve as a scalable and biologically relevant data layer linking prediction and intervention across multiple levels of biological organization. Conclusions:Optical digital twins are no longer merely a conceptual aspiration but are emerging as a practical, measurement-driven infrastructure for precision medicine. The primary challenge is no longer feasibility, but rather the integration, interoperability, validation, and uncertainty quantification of digital twin systems capable of operating safely and at scale. Advances in photonic sensing, computational modeling, and clinical translation position optical digital twins to support real-time, patient-specific clinical decision-making across diagnosis, treatment, monitoring, and prevention.
Significance:Multiphoton fluorescence microscopy is the technique of choice for investigations of thick, highly scattering samples, but is outperformed by single-photon super-resolution techniques in spatial resolving power. Aim:We combine two-photon microscopy with two super-resolution microscopy methods, namely, image scanning microscopy and super-resolution optical fluctuation imaging to overcome the reduction in resolution of laser scanning multiphoton microscopy compared with confocal microscopy. Making use of higher-order cumulants and image deconvolution a resolution better than 100 nm can be achieved. Approach:Two-photon image scanning optical fluctuation imaging is achieved by detecting the descanned signal of fluorescence on a 23-element single photon avalanche detector and analyzing (higher order) cumulants of the temporal evolution of the signals. We test the performance of our method with samples of dispersed quantum dots. We show the applicability of two-photon image scanning optical fluctuation imaging to biological samples with fixed mouse ventral midbrain neurons with quantum dot labeled tubulin. Results:Combining two photon laser scanning microscopy with image scanning microscopy allows to overcome the reduction in resolution caused by the longer wavelength excitation inherent to multiphoton excitation. Analyzing the temporal fluctuations of the signals by calculating cumulants allows to surpass the resolution achieved with conventional confocal imaging of the same fluorophores, and the use of higher-order cumulants and deconvolution allows to achieve a lateral resolution of 75 nm when imaging quantum dots emitting at 625 nm. Conclusions:Combining two-photon microscopy with image scanning microscopy and optical fluctuation imaging allows to achieve a 5-fold improvement in resolution over standard two-photon microscopy, and a 3.5-fold improvement over conventional widefield imaging of the same fluorophores. This represents the first time, to our knowledge, that sub-100 nm imaging is achieved using multiphoton laser scanning microscopy.
SignificanceUltra-widefield scanning laser ophthalmoscopy (UWF SLO), a confocal scanning-based ophthalmic imaging modality widely used in clinical and research ophthalmology, offers an ultra-wide field of view, high resolution, and real-time dynamic imaging. However, multiwavelength imaging-induced transverse chromatic aberration (TCA) is markedly exacerbated in the peripheral retina, causing ghosting artifacts of retinal vessels and other critical anatomical structures. This severely limits image quality and the reliability of downstream quantitative analysis.AimWe aim to correct TCA in multiwavelength UWF SLO images using a subregion-based maximum similarity search registration method, thereby eliminating vascular ghosting artifacts to enhance image quality for reliable clinical analysis.ApproachIn this study, we address the nonuniform TCA between the red and green channels in multiwavelength UWF SLO imaging through three core steps: grid-based subregion sampling, high-precision control point extraction via local optimal matching guided by zero-mean normalized cross-correlation (ZNCC), and global perspective transformation model fitting. The effectiveness of the proposed method was validated on fundus image samples acquired by our in-house-developed UWF SLO system, through both qualitative visual assessment and quantitative evaluation metrics.ResultsThe proposed correction method significantly mitigated TCA in UWF SLO images and greatly improved the spatial alignment of retinal vascular contours between the red and green channels. Validation on 11 fundus images showed a significant enhancement in vascular matching performance: the mean dice similarity coefficient increased by 26.7% relatively (from 0.595±0.060 to 0.754±0.035) and the mean intersection over union increased by 42.5% relatively (from 0.426±0.061 to 0.607±0.044). Paired samples t-test verified highly statistically significant improvements in both metrics (all p<0.001).ConclusionsThe proposed subregion-based maximum similarity search registration method effectively corrects TCA in multiwavelength UWF SLO fundus images, significantly improves image quality, and provides reliable technical support for clinical ophthalmic practice and downstream quantitative fundus image analysis.
Significance:The slit lamp is the gold standard ophthalmic instrument for examining the anterior segment of the eye. Slit lamps require the mechanical stabilization of patients to facilitate alignment and a specialized technician to operate. The growing demand for ophthalmic care and shortages of specialists highlight the need for alternative diagnostic technologies. Aim:We demonstrate an autonomously aligning, contactless slit lamp enabled by a vision-guided robot to acquire anterior segment images without mechanical patient stabilization. Approach:We integrate three subsystems mounted to a collaborative robot arm: a custom optical system for dynamic slit illumination, vision-based sensors, and color cameras. A hierarchical control strategy is employed to maintain alignment. Programmable illumination patterns are projected onto the cornea to replicate slit-lamp examination maneuvers. Results:The device autonomously aligned to the eyes of in vivo human subjects, and we collected color images of the anterior segment during alignment. Dynamic slit illumination compensated for motion with a latency of 86 ms, allowing slit alignment to be maintained during imaging without mechanical stabilization. Conclusions:These results demonstrate the feasibility of autonomous, contactless slit-lamp imaging through the use of robotic alignment and dynamic illumination. Such advancements could expand access to anterior segment diagnostic imaging.
Significance:Retinal endolaser photocoagulation is a widely performed vitreoretinal procedure but currently lacks real-time intraoperative feedback on treatment outcomes. Integrating optical coherence tomography (OCT) directly into the endolaser probe enables co-localized, intraoperative sensing of tissue response during laser delivery, offering the potential to significantly enhance surgical safety and efficacy. Aim:The aim of the study is to design, fabricate, and demonstrate a fiber-optic smart instrument with a 3D nano-printed microlens and instrument-integrated OCT sensing that enables co-localized, concurrent surgical laser delivery and tomographic sensing of tissue alterations during REPC. Approach:A dual-clad fiber was used to co-axially deliver surgical and OCT beams through the outer multi-mode and inner single-mode cores. A 3D nano-printed aspherical microlens ( 300 μ m diameter) at the fiber tip optimized beam shaping for both lasers in the intravitreal cavity. The optical performance of the 23G instrument-integrated OCT (iiOCT) probe was characterized, and ex vivo validation experiments were conducted on porcine eyes. Results:Optical performance assessments demonstrated close alignment between measured and simulated beam profiles, with an OCT spot size of 37.9 μ m at best focus and a surgical laser beam divergence half-angle of 6.0 deg. The probe demonstrated stable power transmission with a mean deviation of 0.35% across repeated pulses. Ex vivo experiments confirmed co-localized, concurrent surgical laser delivery and OCT sensing of retinal tissue alterations. Conclusions:The miniaturized iiOCT endolaser probe combines surgical laser delivery with OCT sensing, achieving adequate performance for the targeted clinical application. This sensor-integrated instrument has the potential to enable intraocular, intraoperative, and quantitative assessment of treatment outcomes. It could support feedback-driven, robotic endolaser photocoagulation, enhancing the safety and efficacy of retinal endolaser surgery.
Significance:Destructive staining and slow subjective visual inspection remain major limitations in rapid evaluation of biopsy specimens during endoscopic ultrasound-guided fine needle biopsy. Therefore, there is a strong need for a label-free optical imaging approach that enables quantitative specimen assessment while preserving specimen integrity. Aim:A multimodal imaging system integrating hyperspectral and color imaging was developed to acquire images of biopsy specimens without staining or sample preparation. Approach:A hyperspectral on-site evaluation workflow was proposed, integrating hyperspectral imaging and color imaging within a modular platform. Hyperspectral data were analyzed using a support-vector-machine-based segmentation framework for specimen assessment with color images providing reference annotations. No washing or staining is required. Results:Validation on 220 pancreatic biopsy specimens demonstrates accurate non-destructive macroscopic visible core evaluation, achieving an F 1 -score of 94.7%. The proposed workflow improves analysis efficiency by 93% compared with conventional rapid on-site evaluation. Conclusions:We demonstrate that integrating hyperspectral imaging with computational analysis enables rapid, quantitative, and label-free specimen evaluation, highlighting the potential of biomedical optical systems for intra-procedural specimen assessment.
Significance:Optical transmission imaging is an inexpensive and noninvasive method of examining biological tissues and structures, but due to strong scattering, its applicability is limited to millimeter-thick samples, thus preventing in vivo studies of centimeter-thick biological organs. Aim:The development of a technique that overcomes the limitation imposed by scattering will significantly expand the scope of the application of transillumination imaging, allowing the examination or diagnosis of biological objects of centimeter thickness, in particular, interphalangeal joints. Approach:The approach used is based on a previously developed method of highly sensitive registration of predominantly ballistic and snake photons when scanning modulated laser radiation. In addition, to obtain a contour image, a technique of selective registration across the profile of a Gaussian laser beam was used, which made it possible to obtain high spatial resolution with a large diameter of a collimated laser beam. An optical clearing agent was used to mitigate the scattering imposed by the skin. Results:The method was tested on a model object and applied to obtain the contour of the interphalangeal joint. The resulting joint contour matches the actual shape, but its width in the image is larger than the actual width. Calculations show that the size of the structure in the image also depends on the intensity and diameter of the laser beam, as well as the extinction coefficient of the sample. It is shown that, given the other parameters, the extinction coefficient can be determined from the joint image space. The effect of the clearing agent on the clarity and contrast of the image is quantitatively analyzed, and a reduction in the scattering coefficient (by 9% to 14%) is estimated. Conclusions:The suggested method has been shown to be efficient for deep-tissue imaging of biological objects up to several centimeters thick, particularly joints, revealing the outline of structural features. We have shown that the use of an optical clearing agent allows reproducible images to be obtained with noticeably higher contrast and clarity, sufficient for preliminary express diagnostics of pathologies.
SignificancePrenatal alcohol exposure is a major cause of neurodevelopmental and growth impairments collectively termed fetal alcohol spectrum disorders (FASD). However, the mechanisms by which ethanol disrupts early embryonic development remain incompletely understood; for example, the potential link between ethanol treatment and biomechanical properties of tissues that lead to defects in development is not understood.AimTo address this gap in knowledge, we used zebrafish (Danio rerio), a powerful vertebrate model with optical transparency and rapid development, to investigate ethanol-induced alterations during embryogenesis.ApproachEmbryos were exposed to varying concentrations of ethanol and assessed using a multimodal imaging approach integrating light sheet fluorescence microscopy, optical coherence tomography, and optical coherence elastography.ResultsThis platform enabled high-resolution, noninvasive, and parallel visualization of structural, molecular, and biomechanical changes. We found that ethanol exposure disrupts Wnt ligand expression in the nervous system, which correlates with a cascade of morphological abnormalities consistently detected across imaging modalities and consistent with alterations to Wnt/β-catenin signaling activity.ConclusionsThese findings highlight the critical role of Wnt/β-catenin signaling in alcohol-induced teratogenesis and underscore the value of zebrafish models and multimodal imaging for advancing our understanding of FASD pathogenesis.
