Your browser doesn't support javascript.
loading
Mostrar: 20 | 50 | 100
Resultados 1 - 20 de 127.503
Filtrar
1.
Sichuan Da Xue Xue Bao Yi Xue Ban ; 52(2): 156-161, 2021 Mar.
Artículo en Chino | MEDLINE | ID: mdl-33829684

RESUMEN

In recent years, with the progress of image processing and network transmission technology, digital pathology (DP) is being more and more extensive applied in clinical practice, and new artificial-intelligence-assisted diagnosis technology based on digital imaging is emerging. Being a widely-used mature field, telepathology is changing the temporal and spatial scope of pathological diagnosis through remote electronic transmission of digital images. Fully digitized pathology departments are realizing the transformation of diagnostic modes and workflow from microscopic diagnosis to digital image computer review, and there have already been successful examples of large-scale fully digitized pathology departments. However, there are still many problems in the implementation of DP, for example, the quality stability and cost of the scanner, the validation of the system, the reengineering of the workflow, the training of pathologists and the change of their perception of DP, which all await further improvement. Although artificial intelligence diagnostic technology is showing great potential, its application in pathological work is still limited to the field of auxiliary diagnostics, and there is still a long way to go to the realization of comprehensive intelligent pathology. The rise of DP will bring about a profound change in the way of how pathological work is done and become a solid foundation for intelligent pathology.


Asunto(s)
Inteligencia Artificial , Telepatología , Procesamiento de Imagen Asistido por Computador
2.
Sichuan Da Xue Xue Bao Yi Xue Ban ; 52(2): 286-292, 2021 Mar.
Artículo en Chino | MEDLINE | ID: mdl-33829704

RESUMEN

Objective: To evaluate the noise reduction effect of deep learning-based reconstruction algorithms in thin-section chest CT images by analyzing images reconstructed with filtered back projection (FBP), adaptive statistical iterative reconstruction (ASIR), and deep learning image reconstruction (DLIR) algorithms. Methods: The chest CT scan raw data of 47 patients were included in this study. Images of 0.625 mm were reconstructed using six reconstruction methods, including FBP, ASIR hybrid reconstruction (ASIR50%, ASIR70%), and deep learning low, medium and high modes (DL-L, DL-M, and DL-H). After the regions of interest were outlined in the aorta, skeletal muscle and lung tissue of each group of images, the CT values, SD values and signal-to-noise ratio (SNR) of the regions of interest were measured, and two radiologists evaluated the image quality. Results: CT values, SD values and SNR of the images obtained by the six reconstruction methods showed statistically significant difference ( P<0.001). There were statistically significant differences in the image quality scores of the six reconstruction methods ( P<0.001). Images reconstruced with DL-H have the lowest noise and the highest overall quality score. Conclusion: The model based on deep learning can effectively reduce the noise of thin-section chest CT images and improve the image quality. Among the three deep-learning models, DL-H showed the best noise reduction effect.


Asunto(s)
Aprendizaje Profundo , Algoritmos , Humanos , Procesamiento de Imagen Asistido por Computador , Dosis de Radiación , Interpretación de Imagen Radiográfica Asistida por Computador , Tomografía Computarizada por Rayos X
3.
Sichuan Da Xue Xue Bao Yi Xue Ban ; 52(2): 293-299, 2021 Mar.
Artículo en Chino | MEDLINE | ID: mdl-33829705

RESUMEN

Objective: To compare the noise reduction performance of conventional filtering and artificial intelligence-based filtering and interpolation (AIFI) and to explore for optimal parameters of applying AIFI in the noise reduction of abdominal magnetic resonance imaging (MRI). Methods: Sixty patients who underwent upper abdominal MRI examination in our hospital were retrospectively included. The raw data of T1-weighted image (T1WI), T2-weighted image (T2WI), and dualecho sequences were reconstructed with two image denoising techniques, conventional filtering and AIFI of different levels of intensity. The difference in objective image quality indicators, peak signal-to-noise ratio (pSNR) and image sharpness, of the different denoising techniques was compared. Two radiologists evaluated the image noise, contrast, sharpness, and overall image quality. Their scores were compared and the interobserver agreement was calculated. Results: Compared with the original images, improvement of varying degrees were shown in the pSNR and the sharpness of the images of the three sequences, T1W1, T2W2, and dual echo sequence, after denoising filtering and AIFI were used (all P<0.05). In addition, compared with conventional filtering, the objective quality scores of the reconstructed images were improved when conventional filtering was combined with AIFI reconstruction methods in T1WI sequence, AIFI level≥3 was used in T2WI and echo1 sequence, and AIFI level≥4 was used in echo2 sequence (all P<0.05). The subjective scores given by the two radiologists for the image noise, contrast, sharpness, and overall image quality in each sequence of conventional filtering reconstruction, AIFI reconstruction (except for AIFI level=1), and two-method combination reconstruction were higher than those of the original images (all P<0.05). However, the image contrast scores were reduced for AIFI level=5. There was good interobserver agreement between the two radiologists (all r>0.75, P<0.05). After multidimensional comparison, the optimal parameters of using AIFI technique for noise reduction in abdominal MRI were conventional filtering+AIFI level=3 in the T1WI sequence and AIFI level=4 in the T2WI and dualecho sequences. Conclusion: AIFI is superior to filtering in imaging denoising at medium and high levels. It is a promising noise reduction technique. The optimal parameters of using AIFI for abdominal MRI are Filtering+AIFI level=3 in the T1WI sequence and AIFI level=4 in T2WI and dualecho sequences.


Asunto(s)
Inteligencia Artificial , Imagen por Resonancia Magnética , Algoritmos , Humanos , Procesamiento de Imagen Asistido por Computador , Estudios Retrospectivos
4.
J Vis Exp ; (169)2021 03 19.
Artículo en Inglés | MEDLINE | ID: mdl-33818563

RESUMEN

Cryogenic electron tomography (cryoET) is a powerful method to study the 3D structure of biological samples in a close-to-native state. Current state-of-the-art cryoET combined with subtomogram averaging analysis enables the high-resolution structural determination of macromolecular complexes that are present in multiple copies within tomographic reconstructions. Tomographic experiments usually require a vast amount of tilt series to be acquired by means of high-end transmission electron microscopes with important operational running-costs. Although the throughput and reliability of automated data acquisition routines have constantly improved over the recent years, the process of selecting regions of interest at which a tilt series will be acquired cannot be easily automated and it still relies on the user's manual input. Therefore, the set-up of a large-scale data collection session is a time-consuming procedure that can considerably reduce the remaining microscope time available for tilt series acquisition. Here, the protocol describes the recently developed implementations based on the SerialEM package and the PyEM software that significantly improve the time-efficiency of grid screening and large-scale tilt series data collection. The presented protocol illustrates how to use SerialEM scripting functionalities to fully automate grid mapping, grid square mapping, and tilt series acquisition. Furthermore, the protocol describes how to use PyEM to select additional acquisition targets in off-line mode after automated data collection is initiated. To illustrate this protocol, its application in the context of high-end data collection of Sars-Cov-2 tilt series is described. The presented pipeline is particularly suited to maximizing the time-efficiency of tomography experiments that require a careful selection of acquisition targets and at the same time a large amount of tilt series to be collected.


Asunto(s)
Microscopía por Crioelectrón/métodos , Tomografía con Microscopio Electrónico/métodos , Procesamiento de Imagen Asistido por Computador/métodos , Sustancias Macromoleculares , Reproducibilidad de los Resultados , Programas Informáticos
5.
Artículo en Inglés | MEDLINE | ID: mdl-33809107

RESUMEN

The purpose of this study is to evaluate the various control parameters of a modeled fast non-local means (FNLM) noise reduction algorithm which can separate color channels in light microscopy (LM) images. To achieve this objective, the tendency of image characteristics with changes in parameters, such as smoothing factors and kernel and search window sizes for the FNLM algorithm, was analyzed. To quantitatively assess image characteristics, the coefficient of variation (COV), blind/referenceless image spatial quality evaluator (BRISQUE), and natural image quality evaluator (NIQE) were employed. When high smoothing factors and large search window sizes were applied, excellent COV and unsatisfactory BRISQUE and NIQE results were obtained. In addition, all three evaluation parameters improved as the kernel size increased. However, the kernel and search window sizes of the FNLM algorithm were shown to be dependent on the image processing time (time resolution). In conclusion, this work has demonstrated that the FNLM algorithm can effectively reduce noise in LM images, and parameter optimization is important to achieve the algorithm's appropriate application.


Asunto(s)
Algoritmos , Microscopía , Procesamiento de Imagen Asistido por Computador
6.
Nat Commun ; 12(1): 1737, 2021 03 19.
Artículo en Inglés | MEDLINE | ID: mdl-33741932

RESUMEN

Innate lymphoid cells (ILCs) emerge in the last few years as important regulators of immune responses and biological processes. Although ILCs are mainly known as tissue-resident cells, their precise localization and interactions with the microenvironment are still unclear. Here we combine a multiplexed immunofluorescence technique and a customized computational, open-source analysis pipeline to unambiguously identify CD127+ ILCs in situ and characterize these cells and their microenvironments. Moreover, we reveal the transcription factor IRF4 as a marker for tonsillar ILC3, and identify conserved stromal landmarks characteristic for ILC localization. We also show that CD127+ ILCs share tissue niches with plasma cells in the tonsil. Our works thus provide a platform for multiparametric histological analysis of ILCs to improve our understanding of ILC biology.


Asunto(s)
Linfocitos/inmunología , Linfocitos/patología , Fenotipo , Análisis Espacial , Algoritmos , Análisis por Conglomerados , Tejido Conectivo/diagnóstico por imagen , Tejido Conectivo/patología , Humanos , Procesamiento de Imagen Asistido por Computador , Inmunidad Innata , Factores Reguladores del Interferón/metabolismo , Subunidad alfa del Receptor de Interleucina-7/metabolismo , Aprendizaje Automático , Tonsila Palatina/diagnóstico por imagen , Tonsila Palatina/patología
7.
Nat Commun ; 12(1): 1744, 2021 03 19.
Artículo en Inglés | MEDLINE | ID: mdl-33741998

RESUMEN

Interferometric scattering microscopy is increasingly employed in biomedical research owing to its extraordinary capability of detecting nano-objects individually through their intrinsic elastic scattering. To significantly improve the signal-to-noise ratio without increasing illumination intensity, we developed photonic resonator interferometric scattering microscopy (PRISM) in which a dielectric photonic crystal (PC) resonator is utilized as the sample substrate. The scattered light is amplified by the PC through resonant near-field enhancement, which then interferes with the <1% transmitted light to create a large intensity contrast. Importantly, the scattered photons assume the wavevectors delineated by PC's photonic band structure, resulting in the ability to utilize a non-immersion objective without significant loss at illumination density as low as 25 W cm-2. An analytical model of the scattering process is discussed, followed by demonstration of virus and protein detection. The results showcase the promise of nanophotonic surfaces in the development of resonance-enhanced interferometric microscopies.


Asunto(s)
Microscopía de Interferencia/instrumentación , Microscopía de Interferencia/métodos , Óptica y Fotónica/instrumentación , Óptica y Fotónica/métodos , Cristalización , Diseño de Equipo , Oro , Procesamiento de Imagen Asistido por Computador , Nanopartículas del Metal , Nanoestructuras , Fotones , Proteínas/aislamiento & purificación , Virión/aislamiento & purificación , Virus/aislamiento & purificación
8.
Biosens Bioelectron ; 180: 113088, 2021 May 15.
Artículo en Inglés | MEDLINE | ID: mdl-33647790

RESUMEN

Serial measurement of a large panel of protein biomarkers near the bedside could provide a promising pathway to transform the critical care of acutely ill patients. However, attaining the combination of high sensitivity and multiplexity with a short assay turnaround poses a formidable technological challenge. Here, the authors develop a rapid, accurate, and highly multiplexed microfluidic digital immunoassay by incorporating machine learning-based autonomous image analysis. The assay has achieved 12-plexed biomarker detection in sample volume <15 µL at concentrations < 5 pg/mL while only requiring a 5-min assay incubation, allowing for all processes from sampling to result to be completed within 40 min. The assay procedure applies both a spatial-spectral microfluidic encoding scheme and an image data analysis algorithm based on machine learning with a convolutional neural network (CNN) for pre-equilibrated single-molecule protein digital counting. This unique approach remarkably reduces errors facing the high-capacity multiplexing of digital immunoassay at low protein concentrations. Longitudinal data obtained for a panel of 12 serum cytokines in human patients receiving chimeric antigen receptor-T (CAR-T) cell therapy reveals the powerful biomarker profiling capability. The assay could also be deployed for near-real-time immune status monitoring of critically ill COVID-19 patients developing cytokine storm syndrome.


Asunto(s)
/inmunología , Citocinas/análisis , Procesamiento de Imagen Asistido por Computador/métodos , Inmunoensayo/métodos , Aprendizaje Automático , Análisis por Micromatrices/métodos , Técnicas Analíticas Microfluídicas/métodos , Síndrome de Liberación de Citoquinas , Humanos , Inmunoterapia Adoptiva , Redes Neurales de la Computación
9.
BMC Bioinformatics ; 22(1): 157, 2021 Mar 25.
Artículo en Inglés | MEDLINE | ID: mdl-33765911

RESUMEN

BACKGROUND: Large-scale biological data sets are often contaminated by noise, which can impede accurate inferences about underlying processes. Such measurement noise can arise from endogenous biological factors like cell cycle and life history variation, and from exogenous technical factors like sample preparation and instrument variation. RESULTS: We describe a general method for automatically reducing noise in large-scale biological data sets. This method uses an interaction network to identify groups of correlated or anti-correlated measurements that can be combined or "filtered" to better recover an underlying biological signal. Similar to the process of denoising an image, a single network filter may be applied to an entire system, or the system may be first decomposed into distinct modules and a different filter applied to each. Applied to synthetic data with known network structure and signal, network filters accurately reduce noise across a wide range of noise levels and structures. Applied to a machine learning task of predicting changes in human protein expression in healthy and cancerous tissues, network filtering prior to training increases accuracy up to 43% compared to using unfiltered data. CONCLUSIONS: Network filters are a general way to denoise biological data and can account for both correlation and anti-correlation between different measurements. Furthermore, we find that partitioning a network prior to filtering can significantly reduce errors in networks with heterogenous data and correlation patterns, and this approach outperforms existing diffusion based methods. Our results on proteomics data indicate the broad potential utility of network filters to applications in systems biology.


Asunto(s)
Procesamiento de Imagen Asistido por Computador , Neoplasias , Algoritmos , Difusión , Humanos , Relación Señal-Ruido
10.
Comput Methods Programs Biomed ; 202: 106004, 2021 Apr.
Artículo en Inglés | MEDLINE | ID: mdl-33662804

RESUMEN

BACKGROUND AND OBJECTIVE: Coronavirus disease 2019 (COVID-19) is a highly contagious virus spreading all around the world. Deep learning has been adopted as an effective technique to aid COVID-19 detection and segmentation from computed tomography (CT) images. The major challenge lies in the inadequate public COVID-19 datasets. Recently, transfer learning has become a widely used technique that leverages the knowledge gained while solving one problem and applying it to a different but related problem. However, it remains unclear whether various non-COVID19 lung lesions could contribute to segmenting COVID-19 infection areas and how to better conduct this transfer procedure. This paper provides a way to understand the transferability of non-COVID19 lung lesions and a better strategy to train a robust deep learning model for COVID-19 infection segmentation. METHODS: Based on a publicly available COVID-19 CT dataset and three public non-COVID19 datasets, we evaluate four transfer learning methods using 3D U-Net as a standard encoder-decoder method. i) We introduce the multi-task learning method to get a multi-lesion pre-trained model for COVID-19 infection. ii) We propose and compare four transfer learning strategies with various performance gains and training time costs. Our proposed Hybrid-encoder Learning strategy introduces a Dedicated-encoder and an Adapted-encoder to extract COVID-19 infection features and general lung lesion features, respectively. An attention-based Selective Fusion unit is designed for dynamic feature selection and aggregation. RESULTS: Experiments show that trained with limited data, proposed Hybrid-encoder strategy based on multi-lesion pre-trained model achieves a mean DSC, NSD, Sensitivity, F1-score, Accuracy and MCC of 0.704, 0.735, 0.682, 0.707, 0.994 and 0.716, respectively, with better genetalization and lower over-fitting risks for segmenting COVID-19 infection. CONCLUSIONS: The results reveal the benefits of transferring knowledge from non-COVID19 lung lesions, and learning from multiple lung lesion datasets can extract more general features, leading to accurate and robust pre-trained models. We further show the capability of the encoder to learn feature representations of lung lesions, which improves segmentation accuracy and facilitates training convergence. In addition, our proposed Hybrid-encoder learning method incorporates transferred lung lesion features from non-COVID19 datasets effectively and achieves significant improvement. These findings promote new insights into transfer learning for COVID-19 CT image segmentation, which can also be further generalized to other medical tasks.


Asunto(s)
Procesamiento de Imagen Asistido por Computador , Pulmón/diagnóstico por imagen , Pulmón/fisiopatología , Tomografía Computarizada por Rayos X , Algoritmos , Bases de Datos Factuales , Humanos
11.
J Vis Exp ; (168)2021 02 17.
Artículo en Inglés | MEDLINE | ID: mdl-33682852

RESUMEN

Temperature control is a recent development that provides an additional degree of freedom to study nanochemistry by liquid cell transmission electron microscopy. In this paper, we describe how to prepare an in situ heating experiment for studying the effect of temperature on the formation of gold nanoparticles driven by radiolysis in water. The protocol of the experiment is fairly simple involving a special liquid cell with uniform heating capabilities up to 100 °C, a liquid-cell TEM holder with flow capabilities and an integrated interface for controlling the temperature. We show that the nucleation and growth mechanisms of gold nanoparticles are drastically impacted by the temperature in liquid cell. Using STEM imaging and nanodiffraction, the evolution of the density, size, shape and atomic structure of the growing nanoparticles are revealed in real time. Automated image processing algorithms are exploited to extract useful quantitative data from video sequences, such as the nucleation and growth rates of nanoparticles. This approach provides new inputs for understanding the complex physico-chemical processes at play during the liquid-phase synthesis of nanomaterials.


Asunto(s)
Nanopartículas del Metal/química , Nanopartículas del Metal/ultraestructura , Microscopía Electrónica de Transmisión , Temperatura , Oro/química , Calefacción , Procesamiento de Imagen Asistido por Computador , Programas Informáticos , Agua/química
12.
J Vis Exp ; (168)2021 02 19.
Artículo en Inglés | MEDLINE | ID: mdl-33682856

RESUMEN

The overall goal of this article is to demonstrate a state-of-the-art ultrahigh field (UHF) magnetic resonance (MR) protocol of the brain at 7.0 Tesla in multiple sclerosis (MS) patients. MS is a chronic inflammatory, demyelinating, neurodegenerative disease that is characterized by white and gray matter lesions. Detection of spatially and temporally disseminated T2-hyperintense lesions by the use of MRI at 1.5 T and 3 T represents a crucial diagnostic tool in clinical practice to establish accurate diagnosis of MS based on the current version of the 2017 McDonald criteria. However, the differentiation of MS lesions from brain white matter lesions of other origins can sometimes be challenging due to their resembling morphology at lower magnetic field strengths (typically 3 T). Ultrahigh field MR (UHF-MR) benefits from increased signal-to-noise ratio and enhanced spatial resolution, both key to superior imaging for more accurate and definitive diagnoses of subtle lesions. Hence, MRI at 7.0 T has shown encouraging results to overcome the challenges of MS differential diagnosis by providing MS-specific neuroimaging markers (e.g., central vein sign, hypointense rim structures and differentiation of MS grey matter lesions). These markers and others can be identified by other MR contrasts other than T1 and T2 (T2*, phase, diffusion) and substantially improve the differentiation of MS lesions from those occurring in other neuroinflammatory conditions such as neuromyelitis optica and Susac syndrome. In this article, we describe our current technical approach to study cerebral white and grey matter lesions in MS patients at 7.0 T using different MR acquisition methods. The up-to-date protocol includes the preparation of the MR setup including the radio-frequency coils customized for UHF-MR, standardized screening, safety and interview procedures with MS patients, patient positioning in the MR scanner and acquisition of dedicated brain scans tailored for examining MS.


Asunto(s)
Imagen por Resonancia Magnética , Esclerosis Múltiple/diagnóstico por imagen , Corteza Cerebral/diagnóstico por imagen , Corteza Cerebral/patología , Humanos , Procesamiento de Imagen Asistido por Computador , Esclerosis Múltiple/patología , Neuroimagen , Programas Informáticos , Sustancia Blanca/diagnóstico por imagen , Sustancia Blanca/patología
13.
J Vis Exp ; (168)2021 02 17.
Artículo en Inglés | MEDLINE | ID: mdl-33682862

RESUMEN

Chronic non-healing wounds, which primarily affect the elderly and diabetic, are a significant area of clinical unmet need. Unfortunately, current chronic wound treatments are inadequate, while available pre-clinical models poorly predict the clinical efficacy of new therapies. Here we describe a high throughput, pre-clinical model to assess multiple aspects of the human skin repair response. Partial thickness wounds were created in human ex vivo skin and cultured across a healing time course. Skin wound biopsies were collected in fixative for the whole-mount staining procedure. Fixed samples were blocked and incubated in primary antibody, with detection achieved via fluorescently conjugated secondary antibody. Wounds were counterstained and imaged via confocal microscopy before calculating percentage wound closure (re-epithelialization) in each biopsy. Applying this protocol, we reveal that 2 mm excisional wounds created in healthy donor skin are fully re-epithelialized by day 4-5 post-wounding. On the contrary, closure rates of diabetic skin wounds are significantly reduced, accompanied by perturbed barrier reformation. Combining human skin wounding with a novel whole-mount staining approach allows a rapid and reproducible method to quantify ex vivo wound repair. Collectively, this protocol provides a valuable human platform to evaluate the effectiveness of potential wound therapies, transforming pre-clinical testing and validation.


Asunto(s)
Modelos Biológicos , Piel/patología , Coloración y Etiquetado , Cicatrización de Heridas , Anciano , Animales , Anticuerpos/metabolismo , Medios de Cultivo , Diabetes Mellitus/patología , Humanos , Procesamiento de Imagen Asistido por Computador , Indicadores y Reactivos , Reproducibilidad de los Resultados , Piel/lesiones , Cicatrización de Heridas/fisiología
14.
Nat Commun ; 12(1): 1550, 2021 03 10.
Artículo en Inglés | MEDLINE | ID: mdl-33692351

RESUMEN

Mapping biological processes in brain tissues requires piecing together numerous histological observations of multiple tissue samples. We present a direct method that generates readouts for a comprehensive panel of biomarkers from serial whole-brain slices, characterizing all major brain cell types, at scales ranging from subcellular compartments, individual cells, local multi-cellular niches, to whole-brain regions from each slice. We use iterative cycles of optimized 10-plex immunostaining with 10-color epifluorescence imaging to accumulate highly enriched image datasets from individual whole-brain slices, from which seamless signal-corrected mosaics are reconstructed. Specific fluorescent signals of interest are isolated computationally, rejecting autofluorescence, imaging noise, cross-channel bleed-through, and cross-labeling. Reliable large-scale cell detection and segmentation are achieved using deep neural networks. Cell phenotyping is performed by analyzing unique biomarker combinations over appropriate subcellular compartments. This approach can accelerate pre-clinical drug evaluation and system-level brain histology studies by simultaneously profiling multiple biological processes in their native anatomical context.


Asunto(s)
Mapeo Encefálico/métodos , Encéfalo/diagnóstico por imagen , Humanos , Procesamiento de Imagen Asistido por Computador , Redes Neurales de la Computación
15.
J Vis Exp ; (169)2021 03 06.
Artículo en Inglés | MEDLINE | ID: mdl-33749674

RESUMEN

The tissue hydrogel delipidation method (CLARITY), originally developed by the Deisseroth laboratory, has been modified and widely used for immunostaining and imaging of thick brain samples. However, this advanced technology has not yet been used for whole-mount retinas. Although the retina is partially transparent, its thickness of approximately 200 µm (in mice) still limits the penetration of antibodies into the deep tissue as well as reducing light penetration for high-resolution imaging. Here, we adapted the CLARITY method for whole-mount mouse retinas by polymerizing them with an acrylamide monomer to form a nanoporous hydrogel and then clearing them in sodium dodecyl sulfate to minimize protein loss and avoid tissue damage. CLARITY-processed retinas were immunostained with antibodies for retinal neurons, glial cells, and synaptic proteins, mounted in a refractive index matching solution, and imaged. Our data demonstrate that CLARITY can improve the quality of standard immunohistochemical staining and imaging for retinal neurons and glial cells in whole-mount preparation. For instance, 3D resolution of fine axon-like and dendritic structures of dopaminergic amacrine cells were much improved by CLARITY. Compared to non-processed whole-mount retinas, CLARITY can reveal immunostaining for synaptic proteins such as postsynaptic density protein 95. Our results show that CLARITY renders the retina more optically transparent after the removal of lipids and preserves fine structures of retinal neurons and their proteins, which can be routinely used for obtaining high-resolution imaging of retinal neurons and their subcellular structures in whole-mount preparation.


Asunto(s)
Retina/metabolismo , Coloración y Etiquetado/métodos , Células Amacrinas/fisiología , Animales , Dendritas/fisiología , Neuronas Dopaminérgicas/fisiología , Procesamiento de Imagen Asistido por Computador , Ratones , Microscopía Confocal/métodos , Proteínas del Tejido Nervioso/metabolismo , Receptores AMPA/metabolismo , Refractometría
16.
Fa Yi Xue Za Zhi ; 37(1): 15-20, 2021 Feb.
Artículo en Inglés, Chino | MEDLINE | ID: mdl-33780179

RESUMEN

Abstract: Objective To explore the feasibility of the CT image reconstruction of laryngeal cartilage and hyoid bone in adult age estimation using data mining methods. Methods The neck thin slice CT scans of 413 individuals aged 18 to <80 years were collected and divided into test set and train set, randomly. According to grading methods such as TURK et al., all samples were graded comprehensively. The process of thyroid cartilage ossification was divided into 6 stages, the process of cricoid cartilage ossification was divided into 5 stages, and the synosteosis between the greater horn of hyoid and hyoid body was divided into 3 stages. Multiple linear regression model, support vector regression model, and Bayesian ridge regression model were developed for adult age estimation by scikit-learn 0.17 machine learning kit (Python language). Leave-one-out cross-validation and the test set were used to further evaluate performance of the models. Results All indicators were moderately or poorly associated with age. The model with the highest accuracy in male age estimation was the support vector regression model, with a mean absolute error of 8.67 years, much higher than the other two models. The model with the highest accuracy in female adult age estimation was the support vector regression model, with a mean absolute error of 12.69 years, but its accuracy differences with the other two models had no statistical significance. Conclusion Data mining technology can improve the accuracy of adult age estimation, but the accuracy of adult age estimation based on laryngeal cartilage and hyoid bone is still not satisfactory, so it should be combined with other indicators in practice.


Asunto(s)
Hueso Hioides , Cartílagos Laríngeos , Adolescente , Adulto , Teorema de Bayes , Niño , Minería de Datos , Femenino , Humanos , Hueso Hioides/diagnóstico por imagen , Procesamiento de Imagen Asistido por Computador , Cartílagos Laríngeos/diagnóstico por imagen , Masculino , Tomografía Computarizada por Rayos X
17.
Nat Commun ; 12(1): 1609, 2021 03 11.
Artículo en Inglés | MEDLINE | ID: mdl-33707455

RESUMEN

Histopathological images are used to characterize complex phenotypes such as tumor stage. Our goal is to associate features of stained tissue images with high-dimensional genomic markers. We use convolutional autoencoders and sparse canonical correlation analysis (CCA) on paired histological images and bulk gene expression to identify subsets of genes whose expression levels in a tissue sample correlate with subsets of morphological features from the corresponding sample image. We apply our approach, ImageCCA, to two TCGA data sets, and find gene sets associated with the structure of the extracellular matrix and cell wall infrastructure, implicating uncharacterized genes in extracellular processes. We find sets of genes associated with specific cell types, including neuronal cells and cells of the immune system. We apply ImageCCA to the GTEx v6 data, and find image features that capture population variation in thyroid and in colon tissues associated with genetic variants (image morphology QTLs, or imQTLs), suggesting that genetic variation regulates population variation in tissue morphological traits.


Asunto(s)
Biología Computacional/métodos , Regulación Neoplásica de la Expresión Génica/genética , Expresión Génica/genética , Neoplasias/patología , Sitios de Carácter Cuantitativo/genética , Proteína BRCA1/genética , Biomarcadores de Tumor/genética , Membrana Celular/genética , Membrana Celular/fisiología , Matriz Extracelular/genética , Matriz Extracelular/fisiología , Humanos , Procesamiento de Imagen Asistido por Computador , Neoplasias/genética , Polimorfismo de Nucleótido Simple/genética
18.
Sensors (Basel) ; 21(4)2021 Feb 19.
Artículo en Inglés | MEDLINE | ID: mdl-33669539

RESUMEN

Colon carcinoma is one of the leading causes of cancer-related death in both men and women. Automatic colorectal polyp segmentation and detection in colonoscopy videos help endoscopists to identify colorectal disease more easily, making it a promising method to prevent colon cancer. In this study, we developed a fully automated pixel-wise polyp segmentation model named A-DenseUNet. The proposed architecture adapts different datasets, adjusting for the unknown depth of the network by sharing multiscale encoding information to the different levels of the decoder side. We also used multiple dilated convolutions with various atrous rates to observe a large field of view without increasing the computational cost and prevent loss of spatial information, which would cause dimensionality reduction. We utilized an attention mechanism to remove noise and inappropriate information, leading to the comprehensive re-establishment of contextual features. Our experiments demonstrated that the proposed architecture achieved significant segmentation results on public datasets. A-DenseUNet achieved a 90% Dice coefficient score on the Kvasir-SEG dataset and a 91% Dice coefficient score on the CVC-612 dataset, both of which were higher than the scores of other deep learning models such as UNet++, ResUNet, U-Net, PraNet, and ResUNet++ for segmenting polyps in colonoscopy images.


Asunto(s)
Colonoscopía , Procesamiento de Imagen Asistido por Computador , Redes Neurales de la Computación , Humanos , Masculino
19.
Sensors (Basel) ; 21(4)2021 Feb 18.
Artículo en Inglés | MEDLINE | ID: mdl-33670827

RESUMEN

Iris segmentation plays an important and significant role in the iris recognition system. The prerequisite for accurate iris recognition is the correctness of iris segmentation. However, the efficiency and robustness of traditional iris segmentation methods are severely challenged in a non-cooperative environment because of unfavorable factors, for instance, occlusion, blur, low resolution, off-axis, motion, and specular reflections. All of the above factors seriously reduce the accuracy of iris segmentation. In this paper, we present a novel iris segmentation algorithm that localizes the outer and inner boundaries of the iris image. We propose a neural network model called "Interleaved Residual U-Net" (IRUNet) for semantic segmentation and iris mask synthesis. The K-means clustering is applied to select saliency points set in order to recover the outer boundary of the iris, whereas the inner border is recovered by selecting another set of saliency points on the inner side of the mask. Experimental results demonstrate that the proposed iris segmentation algorithm can achieve the mean IOU value of 98.9% and 97.7% for inner and outer boundary estimation, respectively, which outperforms the existing approaches on the challenging CASIA-Iris-Thousand database.


Asunto(s)
Algoritmos , Procesamiento de Imagen Asistido por Computador , Iris , Bases de Datos Factuales , Iris/diagnóstico por imagen , Redes Neurales de la Computación
20.
Nat Commun ; 12(1): 1478, 2021 03 05.
Artículo en Inglés | MEDLINE | ID: mdl-33674570

RESUMEN

The recently introduced minimal photon fluxes (MINFLUX) concept pushed the resolution of fluorescence microscopy to molecular dimensions. Initial demonstrations relied on custom made, specialized microscopes, raising the question of the method's general availability. Here, we show that MINFLUX implemented with a standard microscope stand can attain 1-3 nm resolution in three dimensions, rendering fluorescence microscopy with molecule-scale resolution widely applicable. Advances, such as synchronized electro-optical and galvanometric beam steering and a stabilization that locks the sample position to sub-nanometer precision with respect to the stand, ensure nanometer-precise and accurate real-time localization of individually activated fluorophores. In our MINFLUX imaging of cell- and neurobiological samples, ~800 detected photons suffice to attain a localization precision of 2.2 nm, whereas ~2500 photons yield precisions <1 nm (standard deviation). We further demonstrate 3D imaging with localization precision of ~2.4 nm in the focal plane and ~1.9 nm along the optic axis. Localizing with a precision of <20 nm within ~100 µs, we establish this spatio-temporal resolution in single fluorophore tracking and apply it to the diffusion of single labeled lipids in lipid-bilayer model membranes.


Asunto(s)
Imagenología Tridimensional/instrumentación , Imagenología Tridimensional/métodos , Microscopía Fluorescente/instrumentación , Microscopía Fluorescente/métodos , Difusión , Diseño de Equipo , Fluorescencia , Colorantes Fluorescentes , Procesamiento de Imagen Asistido por Computador , Fotones
SELECCIÓN DE REFERENCIAS
DETALLE DE LA BÚSQUEDA
...