Biomedical optics express最新文献

Segmentation and analysis of microscopic optical images is a fundamental task in the field of biomedicine. However, efficiently, accurately, and robustly segmenting regions of interest in these images presents a significant challenge. The segment anything model (SAM) has demonstrated remarkable generalization capabilities in natural image segmentation tasks, revealing its potential for segmenting microscopic optical images. In this study, we propose Brain-SAM, a general automatic segmentation model based on SAM for the automatic segmentation of microscopic optical images. Specifically, we introduce an automatic prompt encoder to enable high-throughput automated segmentation of these images. Additionally, we propose a segmentation optimizer to further enhance the model's segmentation performance. Testing on eight benchmark datasets, representing common scenarios in microscopic optical image segmentation, shows that Brain-SAM outperforms specialized segmentation models in the vast majority of segmentation tasks. Notably, on the Brain, Tek and Lectin3d datasets, Brain-SAM achieved IoU scores of 98.07%, 93.13% and 88.49% respectively, along with Dice scores of 99.03%, 96.44% and 93.89%. Moreover, we provide a series of rich, publicly available brain science image datasets created using fluorescence microscopic optical tomography (fMOST) technology.

Accurate identification of target lung segments during thoracoscopic surgery is critical for successful lung cancer resections but remains challenging with conventional thoracoscopic imaging techniques. We present a real-time statistical gating method that leverages the single-shot laser speckle pattern generated by an 840 nm laser to isolate the blood-scattered speckle component in lung tissue. This enables background-free measurement of hemoglobin's absorption, thereby markedly improving the sensitivity of tissue oxygen saturation detection. By exploiting differences in speckle decorrelation time, our approach reconstructs blood-scattered intensity images from single-frame thoracoscopic captures in real-time. Clinical trials demonstrated a 2.05-fold increase in boundary slope steepness and a 220% improvement in the mean absolute derivative metric, enabling precise differentiation of the segment's boundary during the inflation-deflation procedure in lung segmentectomy. For novice surgeons, manual segmentation accuracy improved from 0.78 to 0.92 (standard segmentectomy) and from 0.82 to 0.89 (rapid segmentectomy) in terms of DICE coefficients. Compatible with standard thoracoscopic systems, our method offers real-time, high-sensitivity visualization of blood absorption dynamics, enhancing surgical precision and reducing operative time. As a computational imaging method, the proposed statistical gating method can be seamlessly integrated into existing thoracoscopic systems by adding near-infrared laser illumination and an embedded GPU core. This highlights its potential for revolutionizing lung cancer surgeries.

Contemporary approaches for multiple optical micro-manipulation typically involve careful pre-engineering of the laser beam shape. In various biomedical and microfluidic scenarios, especially those necessitating unconventional specimen chambers, there is a demand for controlling the collection of micro-objects near fluid-fluid interfaces. For many of these cases, a regular array of trap sites as well as tight confinement are not essential. For such applications near interfaces, we expand on the concept of speckle tweezers (ST), which incorporate randomly distributed light fields for quasi-2D optical manipulation. The proposed technique is demonstrated experimentally by applying ST to govern the movement of polystyrene micro-particles near water-oil and water-air interfaces. The efficacy of the method is validated through the temporal characterization of micro-particle motions, and the confinement of the micro-particles near the interfaces is verified using digital holographic microscopy. However, the methodology has the potential for applications in living cell manipulation, soft functional matter creation, and various industrial processes.

The Editor-in-Chief and Deputy Editor of Biomedical Optics Express announce the award prize for the best paper published in the Journal between 2022 and 2024.

Frequency-domain near-infrared spectroscopy (FD-NIRS) is a noninvasive in vivo sensing and imaging technique that is used to quantify tissue composition and oxygen metabolism in the brain, muscle, and other tissues. However, the size and complexity of FD-NIRS instrumentation have largely limited its use to laboratory and clinical research settings. To expand the use of FD-NIRS into continuous monitoring applications and in naturalistic environments, we report a novel real-time multi-frequency (50-350 MHz) wearable FD-NIRS system based on a custom application-specific integrated circuit (ASIC). The wearable device includes 685 nm and 850 nm laser diodes and a silicon photomultiplier (SiPM) detector. The system has an optical property accuracy of 0.0007 mm^-1 for absorption and 0.08 mm^-1 for reduced scattering at a 14.7 Hz measurement rate, evaluated using tissue-simulating phantoms, and is capable of capturing single-frequency measurements up to 1 kHz with both wavelengths. Without a battery, it weighs 37 grams and measures less than 7 x 3 x 3 cm in size. For proof-of-concept, we demonstrate measurement of an arteriovenous occlusion of the human forearm. Overall, this work demonstrates the feasibility of quantitative FD-NIRS tissue optical spectroscopy for wearable health monitoring.

Estimating the scattering coefficient μ _s and scattering anisotropy factor g of the corneal tissue is currently limited to methods utilizing double integrating spheres or spectroscopic techniques, prohibiting such corneal tissue evaluation from being performed in a clinical setting. This paper presents a new concept of statistical matching between a given corneal optical coherence tomography (OCT) scan and a set of multi-reference phantom OCT B-scans, suitably simulated using the Monte Carlo method. The statistical matching that exploits the information present in the speckle includes an ensemble of distance measures. Using a set of OCT scans from 11 porcine eyeballs, for which epithelium was removed, it is demonstrated that the proposed statistical matching approach leads to a precise estimation of the optical properties of corneal stroma. The group mean of the scattering coefficient and the scattering anisotropy factor for the porcine stroma were ${\overline{\mu }}_s=0.146\pm 0.020$ mm^-1 and $\overline{g}=0.893\pm 0.021$ , respectively. These estimates match previously reported values established with other methods. The proposed approach of utilizing information present in the speckle can be readily extended to in vivo OCT imaging of human corneal tissue.

We present dynamic spectrally encoded confocal microscopy (D-SECM), a high-speed, label-free imaging modality for real-time visualization of subcellular metabolic activity. Sequential, optically sectioned SECM images, with a lateral resolution of 0.62 µm-0.88 µm, were acquired at 100 fps by a modified swept source SECM system, stabilized by fiber-Bragg grating swept source synchronization. D-SECM images are created through real-time pixelwise power spectra mapping into frequency-dependent RGB channels, enabled by GPU-accelerated parallel processing. Validation on freshly excised mouse liver tissue demonstrates that metabolic contrast is detectable with exposure durations as short as 40 ms in the liver. Moreover, volume-prioritized scanning enables three-dimensional D-SECM images to be acquired within a second. In addition to the liver, D-SECM has also been demonstrated in multiple organs, including the lung, thyroid, kidney, and pancreas. These capabilities position D-SECM as a powerful and scalable platform for high-speed, label-free imaging of metabolic dynamics, with unique potential for both basic research and real-time clinical applications.

Label-free cell imaging using phase-sensitive optical interferometric microscopy enables noninvasive observation of living cells, but it often suffers from low imaging specificity and limited spatial resolution, particularly in the axial direction. In this study, we present a label-free method for high-resolution 3D chromatin imaging by leveraging rapidly fluctuating scattering signals arising from native biomolecular motions, captured using a high-speed and highly sensitive interferometric microscope. Optical transmission images of live cell nuclei are recorded at 1000 frames per second, and temporal variance maps are computed from these recordings. Deep learning models are then trained to map the label-free dynamics data to confocal fluorescence images of chromatin. Our results demonstrate that the resulting dynamics maps resolve fine subnuclear structures, including nucleoli and nuclear speckles-the latter being especially difficult to detect using conventional phase microscopy. Notably, the use of second-order temporal statistics leads to significantly enhanced axial resolution, enabling effective 3D imaging of chromatin architecture. This work highlights the potential of temporal signal analysis in fast, label-free optical interferometric microscopy and paves the way for broader applications in high-resolution, label-free imaging of dynamic biological structures.

[This corrects the article on p. 3593 in vol. 13, PMID: 35781949.].

Here, we propose optomechanical devices for steering anesthetized animals during retinal imaging and/or stimulation with stationary ophthalmoscopes. Simple operating procedures ensure that the entrance pupil of the eye remains centered on the exit pupil of the ophthalmoscope during steering, to avoid vignetting. The devices, built with commercially available manual linear stages and motorized rotating devices, can be used to capture image sequences for tiling, as is often done in microscopy. This automated steering system, demonstrated here in mice, is applicable to other animal species and imaging modalities, as well as explanted eyes. The use of these devices can reduce imaging time and retinal light exposure, both of which are important when using ophthalmoscopes with small fields of view, such as adaptive optics ophthalmoscopes, while also improving animal welfare.