Significance Most biological tissue is opaque, and the inside fine structures are invisible even with a microscope. Optical clearing methods can reduce scattering enable us to “look” samples. Current optical suffer from large time consumption, distortion, fluorescence quenching, which greatly limit their applications. Here, we describe nontoxic ultrafast method (FOCM), could clarify 300-µm-thick brain slices within 2 min without morphology distortion. The be preserved up 86% after 11 d....

Abstract Current liver fibrosis scoring by computer-assisted image analytics is not fully automated as it requires manual preprocessing (segmentation and feature extraction) typically based on domain knowledge in pathology. Deep learning-based algorithms can potentially classify these images without the need for through learning from a large dataset of images. We investigated performance classification models built using deep algorithm pre-trained multiple sources to score compared them...

Artificial intelligence can facilitate clinical decision making by considering massive amounts of medical imaging data.Various algorithms have been implemented for different applications.Accurate diagnosis and treatment require reliable interpretable data.For pancreatic tumor diagnosis, only 58.5% images from the First Affiliated Hospital Second Hospital, Zhejiang University School Medicine are used, increasing labor time costs to manually filter out not directly used diagnostic...

Optical microscopy has so far been restricted to superficial layers, leaving many important biological questions unanswered. Random scattering causes the ballistic focus, which is conventionally used for image formation, decay exponentially with depth. imaging beyond regime demonstrated by hybrid techniques that combine light deeper penetration capability of sound waves. Deep inside highly media, focus dimensions restrict resolutions. Here we show iteratively focusing into an ultrasound via...

We present a learning-based Shack-Hartmann wavefront sensor (SHWS) to achieve the high-order aberration detection without image segmentation or centroid positioning. Zernike coefficient amplitudes of aberrations measured from biological samples are referred and expanded generate training datasets. With one SHWS pattern inputted, up 120th modes could be predicted within 10.9 ms with 95.56% model accuracy by personal computer. The statistical experimental results show that compared traditional...

Non-invasive, real-time imaging and deep focus into tissue are in high demand biomedical research. However, the aberration that is introduced by refractive index inhomogeneity of biological hinders way forward. A rapid focusing with sensor-less corrections, based on machine learning, demonstrated this paper. The proposed method applies Convolutional Neural Network (CNN), which can rapidly calculate low-order aberrations from point spread function images Zernike modes after training. results...

The conventional Shack-Hartmann wavefront sensor (SHWS) requires slope measurements of every micro-lens for reconstruction. In this Letter, we applied deep learning on the SHWS to directly predict distributions without measurements. results show that our method could provide a lower root mean square error in high detection speed. performance proposed is also evaluated challenging wavefronts, while approaches perform insufficiently. This Letter provides new approach, best knowledge, direct...

Aberrations and random scattering severely limit optical imaging in deep tissue. Adaptive optics can principle drastically extend the penetration depth improve image quality. However, for media a large number of spatial modes need to be measured controlled restore diffraction limited focus. Here, we present parallel wavefront optimization method using backscattered light as feedback. Spatial confinement feedback signal is realized with confocal pinhole coherence gating. We show simulations...

Abstract The amygdala, one of the most studied brain structures, integrates brain-wide heterogeneous inputs and governs multidimensional outputs to control diverse behaviors central survival, yet how amygdalar input-output neuronal circuits are organized remains unclear. Using a simplified cell-type- projection-specific retrograde transsynaptic tracing technique, we scrutinized afferent four major output groups in basolateral complex (BLA) that project bed nucleus stria terminals (BNST),...

Determining the nature of orbital tumors is challenging for current imaging interpretation methods, which hinders timely treatment. This study aimed to propose an end-to-end deep learning system automatically diagnose tumors. A multi-center dataset 602 non-contrast-enhanced computed tomography (CT) images were prepared. After image annotation and preprocessing, CT used train test (DL) model following two stages: tumor segmentation classification. The performance on testing set was compared...

Fluorescence images obtained with optical microscopes intrinsically suffer from blur and noise, which can be partially reversed by the deconvolution process. However, process is ill-conditioned, leading to a trade-off between detail preservation noise suppression. Here, we develop 3D-FUDIP fully decouple into two parts: deblurring denoising, achieving an 8-fold improvement in spatial resolution. By adopting Poisson model, obeys quantum nature of photons, our successfully applied various...

The current study aims to investigate the effects of matrine on JAK-STAT signaling transduction pathways in bleomycin (BLM)-induced pulmonary fibrosis (PF) and explore its action mechanism. A total 72 male C57BL/6 mice were randomized into control, model, treatment groups. PF models established by instilling BLM intratracheally. group was given daily through gastric lavage. Six sacrificed each at 3, 7, 14, 28 days. lung tissues observed using hematoxylin-eosin staining. expression JAK,...

Abstract Visualization of axons and dendritic spines is crucial in neuroscience research. However, traditional microscopy limited by diffraction-limited resolution shallow imaging depth, making it difficult to study neuronal dynamics. Two-photon multifocal structured illumination (2P-MSIM) provides super-resolution along with a reasonably good penetration, but vulnerable optical aberrations deep tissues. Herein we present novel non-inertial scanning 2P-MSIM system incorporated adaptive...

Photoacoustic imaging relies on diffused photons for optical contrast and diffracted ultrasound high resolution. As a tomographic modality, often an inverse problem of acoustic diffraction needs to be solved reconstruct photoacoustic image. The is complicated by the fact that properties, including speed sound distribution, in image field view are unknown. During reconstruction, subtle changes ray path may accumulate give rise noticeable blurring Thus, addition detection bandwidth, inaccurate...

Glaucoma is the main cause of irreversible blindness worldwide. However, diagnosis and treatment glaucoma remain difficult because lack an effective grading measure. In this study, we aimed to propose artificial intelligence system provide adequate assessment patients. A total 16,356 visual fields (VFs) measured by Octopus perimeters Humphrey Field Analyzer (HFA) were collected, from three hospitals in China public Harvard database. We developed a fine-grained deep learning system, named...

Glaucoma is the leading cause of irreversible blindness, and early detection timely treatment are essential for glaucoma management. However, due to interindividual variability in characteristics onset, a single feature not yet sufficient monitoring progression isolation. There an urgent need develop more comprehensive diagnostic methods with higher accuracy. In this study, we proposed multi- deep learning (MFDL) system based on intraocular pressure (IOP), color fundus photograph (CFP)...

Abstract Flexible integrated photonics is an essential technology for emerging applications, including flexible optical interconnects, optogenetic stimulation, and implantable conformal sensing. Here, a novel universal route fabricating photonic components with high‐refractive‐index contrast reported. Central to such unique method the utilization of germanium oxide (GeO) as sacrificial layer releasing nanostructures from rigid substrates substrates. Various high‐quality inorganic materials...

Two‐photon microscopy (2PM) is one of the most widely used tools for in vivo deep tissue imaging. However, spatial resolution and penetration depth are still limited due to strong scattering background. Here we demonstrate a two‐photon focal modulation microscopy. By utilizing demodulation techniques, background rejection capability enhanced, thus imaging improved. Compared with 2PM, transverse increased by 70%, while axial 2‐fold. Furthermore, when applied conventional 2PM mode, it can...

The doughnut-shaped beam has been widely applied in the field of super-resolution microscopic imaging, micro-nanostructure lithography, ultra-high-density storage, and laser trapping. However, how to maintain focus inside scattering medium becomes a challenge, due wavefront aberrations. Here we demonstrate machine learning based adaptive optics method recover with high speed. In our method, relationship between distorted intensity point spread function coefficients first 15 Zernike modes for...

A diffraction analysis is presented for image formation in confocal microscopy using the divided aperture technique, which uses two D-shaped apertures (also called specular microscopy). The effects of increasing width a divider, that separates D shapes, are investigated. As increased, resolution degrades. efficiency singly-scattered light rejection not improved with increased width.

We show that the focal modulation microscopy (FMM), which combines a spatial phase modulator with confocal microscopy, results in an improvement resolution. This technique was introduced to increase imaging depth into tissue and rejection of background from thick scattering object. A theory for image formation FMM is presented, effects detecting in-phase modulated fluorescence signal are discussed. Compared conventional width point-spread function improved by 16.4%. When applied saturable...