RESUMEN
PURPOSE: To achieve an accurate diagnosis of idiopathic epiretinal membrane (iERM) through analyzing retinal blood vessel oxygen saturation (SO2) and vascular morphological features in fundus images. METHODS: Dual-modal fundus camera was used to obtain color fundus image, 570-nm, and 610-nm images. As iERM affects the macular area, a macular-centered semicircle area as region of interest (MROI) was selected and analyzed SO2 and vascular morphologies in it. Eventually, random forest (RF) and support vector machine (SVM) were as classifiers to diagnose iERM patients. RESULTS: The arterial and venous SO2 levels of the iERM group were significantly higher than that of the control group. There were significant differences in the vessel density and fractal dimension on the artery for vascular morphology, while the tortuosity had a significant difference in the vein. By feeding the SO2 and the vascular morphological features into classifiers, an accuracy of over 82% was obtained, which is significantly better than the two separate inputs for classification. CONCLUSION: Significant differences in SO2 and vascular morphologies between control and iERM groups confirmed that the occurrence of iERM may affect blood supply and vascular structures. Benefiting from the dual-modal fundus camera and machine learning models, accurate judgments can be made on fundus images. Extensive experiments proved the importance of blood vessel SO2 and vascular morphologies for diagnosis, which is of great significance for clinical screening.
Asunto(s)
Membrana Epirretinal , Humanos , Membrana Epirretinal/diagnóstico , Saturación de Oxígeno , Fondo de Ojo , Vasos Retinianos/diagnóstico por imagen , Angiografía con Fluoresceína/métodos , OxígenoRESUMEN
BACKGROUND: To define a new quantitative grading criterion for retinal haemorrhages in term newborns based on the segmentation results of a deep convolutional neural network. METHODS: We constructed a dataset of 1543 retina images acquired from 847 term newborns, and developed a deep convolutional neural network to segment retinal haemorrhages, blood vessels and optic discs and locate the macular region. Based on the ratio of areas of retinal haemorrhage to optic disc, and the location of retinal haemorrhages relative to the macular region, we defined a new criterion to grade the degree of retinal haemorrhages in term newborns. RESULTS: The F1 scores of the proposed network for segmenting retinal haemorrhages, blood vessels and optic discs were 0.84, 0.73 and 0.94, respectively. Compared with two commonly used retinal haemorrhage grading criteria, this new method is more accurate, objective and quantitative, with the relative location of the retinal haemorrhages to the macula as an important factor. CONCLUSIONS: Based on a deep convolutional neural network, we can segment retinal haemorrhages, blood vessels and optic disc with high accuracy. The proposed grading criterion considers not only the area of the haemorrhages but also the locations relative to the macular region. It provides a more objective and comprehensive evaluation criterion. The developed deep convolutional neural network offers an end-to-end solution that can assist doctors to grade retinal haemorrhages in term newborns.
Asunto(s)
Procesamiento de Imagen Asistido por Computador/métodos , Redes Neurales de la Computación , Hemorragia Retiniana/clasificación , Hemorragia Retiniana/diagnóstico por imagen , Aprendizaje Profundo , Humanos , Recién Nacido , Disco Óptico/patología , Vasos Retinianos/patología , Nacimiento a TérminoRESUMEN
Retinal vessel oxygen supply is important for retinal tissue metabolism. Commonly used retinal vessel oximetry devices are based on dual-wavelength spectral measurement of oxyhemoglobin and deoxyhemoglobin. However, there is no traceable standard for reliable calibration of these devices. In this study, we developed a fundus-simulating phantom that closely mimicked the optical properties of human fundus tissues. Microchannels of precisely controlled topological structures were produced by soft lithography to simulate the retinal vasculature. Optical properties of the phantom were adjusted by adding scattering and absorption agents to simulate different concentrations of fundus pigments. The developed phantom was used to calibrate the linear correlation between oxygen saturation (SO2) level and optical density ratio in a dual-wavelength oximetry device. The obtained calibration factors were used to calculate the retinal vessel SO2 in both eyes of five volunteers aged between 24 and 27 years old. The test results showed that the mean arterial and venous SO2 levels after phantom calibration were coincident with those after empirical value calibration, indicating the potential clinical utility of the produced phantom as a calibration standard.
RESUMEN
Postnatal skeletal muscle mass is regulated by the balance between anabolic protein synthesis and catabolic protein degradation, and muscle atrophy occurs when protein homeostasis is disrupted. Autophagy has emerged as critical in clearing dysfunctional organelles and thus in regulating protein turnover. Here we show that endolysosomal two-pore channel subtype 2 (TPC2) contributes to autophagy signaling and protein homeostasis in skeletal muscle. Muscles derived from Tpcn2(-/-) mice exhibit an atrophic phenotype with exacerbated autophagy under starvation. Compared with wild types, animals lacking TPC2 demonstrated an enhanced autophagy flux characterized by increased accumulation of autophagosomes upon combined stress induction by starvation and colchicine treatment. In addition, deletion of TPC2 in muscle caused aberrant lysosomal pH homeostasis and reduced lysosomal protease activity. Association between mammalian target of rapamycin and TPC2 was detected in skeletal muscle, allowing for appropriate adjustments to cellular metabolic states and subsequent execution of autophagy. TPC2 therefore impacts mammalian target of rapamycin reactivation during the process of autophagy and contributes to maintenance of muscle homeostasis.
Asunto(s)
Autofagia , Canales de Calcio/metabolismo , Músculo Esquelético/metabolismo , Transducción de Señal , Animales , Canales de Calcio/genética , Homeostasis , Concentración de Iones de Hidrógeno , Lisosomas/metabolismo , Lisosomas/ultraestructura , Masculino , Ratones , Ratones Endogámicos C57BL , Músculo Esquelético/patología , Péptido Hidrolasas/metabolismo , Fagosomas/metabolismo , Fagosomas/ultraestructura , Estrés Fisiológico , Serina-Treonina Quinasas TOR/metabolismoRESUMEN
Cell membrane repair is an important aspect of physiology, and disruption of this process can result in pathophysiology in a number of different tissues, including wound healing, chronic ulcer and scarring. We have previously identified a novel tripartite motif family protein, MG53, as an essential component of the cell membrane repair machinery. Here we report the functional role of MG53 in the modulation of wound healing and scarring. Although MG53 is absent from keratinocytes and fibroblasts, remarkable defects in skin architecture and collagen overproduction are observed in mg53(-/-) mice, and these animals display delayed wound healing and abnormal scarring. Recombinant human MG53 (rhMG53) protein, encapsulated in a hydrogel formulation, facilitates wound healing and prevents scarring in rodent models of dermal injuries. An in vitro study shows that rhMG53 protects against acute injury to keratinocytes and facilitates the migration of fibroblasts in response to scratch wounding. During fibrotic remodeling, rhMG53 interferes with TGF-ß-dependent activation of myofibroblast differentiation. The resulting down-regulation of α smooth muscle actin and extracellular matrix proteins contributes to reduced scarring. Overall, these studies establish a trifunctional role for MG53 as a facilitator of rapid injury repair, a mediator of cell migration, and a modulator of myofibroblast differentiation during wound healing. Targeting the functional interaction between MG53 and TGF-ß signaling may present a potentially effective means for promoting scarless wound healing.
Asunto(s)
Proteínas Portadoras/fisiología , Membrana Celular/metabolismo , Proteínas Musculares/fisiología , Proteínas de Transporte Vesicular/fisiología , Cicatrización de Heridas/fisiología , Células 3T3 , Actinas/metabolismo , Animales , Diferenciación Celular , Movimiento Celular , Cicatriz/patología , Colágeno Tipo I/metabolismo , Fibroblastos/citología , Fibronectinas/metabolismo , Fibrosis/patología , Regulación de la Expresión Génica , Humanos , Hidrogeles/química , Queratinocitos/metabolismo , Proteínas de la Membrana , Ratones , Músculo Liso/metabolismo , Miofibroblastos/metabolismo , Conejos , Ratas , Ratas Sprague-Dawley , Proteínas Recombinantes/metabolismo , Piel/patología , Proteínas de Motivos TripartitosRESUMEN
BACKGROUND: The cardiac cytoskeleton plays key roles in maintaining myocyte structural integrity in health and disease. In fact, human mutations in cardiac cytoskeletal elements are tightly linked to cardiac pathologies, including myopathies, aortopathies, and dystrophies. Conversely, the link between cytoskeletal protein dysfunction and cardiac electric activity is not well understood and often overlooked in the cardiac arrhythmia field. METHODS AND RESULTS: Here, we uncover a new mechanism for the regulation of cardiac membrane excitability. We report that ßII spectrin, an actin-associated molecule, is essential for the posttranslational targeting and localization of critical membrane proteins in heart. ßII spectrin recruits ankyrin-B to the cardiac dyad, and a novel human mutation in the ankyrin-B gene disrupts the ankyrin-B/ßII spectrin interaction, leading to severe human arrhythmia phenotypes. Mice lacking cardiac ßII spectrin display lethal arrhythmias, aberrant electric and calcium handling phenotypes, and abnormal expression/localization of cardiac membrane proteins. Mechanistically, ßII spectrin regulates the localization of cytoskeletal and plasma membrane/sarcoplasmic reticulum protein complexes, including the Na/Ca exchanger, ryanodine receptor 2, ankyrin-B, actin, and αII spectrin. Finally, we observe accelerated heart failure phenotypes in ßII spectrin-deficient mice. CONCLUSIONS: Our findings identify ßII spectrin as critical for normal myocyte electric activity, link this molecule to human disease, and provide new insight into the mechanisms underlying cardiac myocyte biology.
Asunto(s)
Arritmias Cardíacas/patología , Arritmias Cardíacas/fisiopatología , Citoesqueleto/fisiología , Miocitos Cardíacos/patología , Miocitos Cardíacos/fisiología , Espectrina/fisiología , Secuencia de Aminoácidos , Animales , Ancirinas/genética , Ancirinas/fisiología , Arritmias Cardíacas/genética , Proteínas Portadoras/genética , Proteínas Portadoras/fisiología , Modelos Animales de Enfermedad , Insuficiencia Cardíaca/genética , Insuficiencia Cardíaca/patología , Insuficiencia Cardíaca/fisiopatología , Humanos , Proteínas de la Membrana/fisiología , Ratones , Ratones Noqueados , Proteínas de Microfilamentos/deficiencia , Proteínas de Microfilamentos/genética , Proteínas de Microfilamentos/fisiología , Microtúbulos/fisiología , Datos de Secuencia Molecular , Mutación/genética , Fenotipo , Espectrina/análisis , Espectrina/químicaRESUMEN
Single-molecule localization microscopy achieves sub-diffraction-limit resolution by localizing a sparse subset of stochastically activated emitters in each frame. Its temporal resolution is limited by the maximal emitter density that can be handled by the image reconstruction algorithms. Multiple algorithms have been developed to accurately locate the emitters even when they have significant overlaps. Currently, compressive-sensing-based algorithm (CSSTORM) achieves the highest emitter density. However, CSSTORM is extremely computationally expensive, which limits its practical application. Here, we develop a new algorithm (MempSTORM) based on two-dimensional spectrum analysis. With the same localization accuracy and recall rate, MempSTORM is 100 times faster than CSSTORM with â(1)-homotopy. In addition, MempSTORM can be implemented on a GPU for parallelism, which can further increase its computational speed and make it possible for online super-resolution reconstruction of high-density emitters.
Asunto(s)
Algoritmos , Procesamiento de Imagen Asistido por Computador/métodos , MicroscopíaRESUMEN
Myxococcus xanthus is a model organism for studying bacterial social behaviors due to its ability to form complex multi-cellular structures. Knowledge of M. xanthus surface gliding motility and the mechanisms that coordinated it are critically important to our understanding of collective cell behaviors. Although the mechanism of gliding motility is still under investigation, recent experiments suggest that there are two possible mechanisms underlying force production for cell motility: the focal adhesion mechanism and the helical rotor mechanism, which differ in the biophysics of the cell-substrate interactions. Whereas the focal adhesion model predicts an elastic coupling, the helical rotor model predicts a viscous coupling. Using a combination of computational modeling, imaging, and force microscopy, we find evidence for elastic coupling in support of the focal adhesion model. Using a biophysical model of the M. xanthus cell, we investigated how the mechanical interactions between cells are affected by interactions with the substrate. Comparison of modeling results with experimental data for cell-cell collision events pointed to a strong, elastic attachment between the cell and substrate. These results are robust to variations in the mechanical and geometrical parameters of the model. We then directly measured the motor-substrate coupling by monitoring the motion of optically trapped beads and find that motor velocity decreases exponentially with opposing load. At high loads, motor velocity approaches zero velocity asymptotically and motors remain bound to beads indicating a strong, elastic attachment.
Asunto(s)
Adhesión Bacteriana/fisiología , Proteínas Bacterianas/fisiología , Adhesiones Focales/fisiología , Modelos Biológicos , Proteínas Motoras Moleculares/fisiología , Myxococcus xanthus/fisiología , Simulación por Computador , Módulo de Elasticidad/fisiología , Fricción , Movimiento (Física) , Myxococcus xanthus/citología , ViscosidadRESUMEN
One key factor that limits resolution of single-molecule superresolution microscopy relates to the localization accuracy of the activated emitters, which is usually deteriorated by two factors. One originates from the background noise due to out-of-focus signals, sample auto-fluorescence, and camera acquisition noise; and the other is due to the low photon count of emitters at a single frame. With fast acquisition rate, the activated emitters can last multiple frames before they transiently switch off or permanently bleach. Effectively incorporating the temporal information of these emitters is critical to improve the spatial resolution. However, majority of the existing reconstruction algorithms locate the emitters frame by frame, discarding or underusing the temporal information. Here we present a new image reconstruction algorithm based on tracklets, short trajectories of the same objects. We improve the localization accuracy by associating the same emitters from multiple frames to form tracklets and by aggregating signals to enhance the signal to noise ratio. We also introduce a weighted mean-shift algorithm (WMS) to automatically detect the number of modes (emitters) in overlapping regions of tracklets so that not only well-separated single emitters but also individual emitters within multi-emitter groups can be identified and tracked. In combination with a maximum likelihood estimator method (MLE), we are able to resolve low to medium density of overlapping emitters with improved localization accuracy. We evaluate the performance of our method with both synthetic and experimental data, and show that the tracklet-based reconstruction is superior in localization accuracy, particularly for weak signals embedded in a strong background. Using this method, for the first time, we resolve the transverse tubule structure of the mammalian skeletal muscle.
Asunto(s)
Algoritmos , Procesamiento de Imagen Asistido por Computador , Músculo Esquelético/citología , Nanotecnología/métodos , Fotones , Animales , Células Cultivadas , Relación Señal-RuidoRESUMEN
Protein-directed intracellular transport has not been observed in bacteria despite the existence of dynamic protein localization and a complex cytoskeleton. However, protein trafficking has clear potential uses for important cellular processes such as growth, development, chromosome segregation, and motility. Conflicting models have been proposed to explain Myxococcus xanthus motility on solid surfaces, some favoring secretion engines at the rear of cells and others evoking an unknown class of molecular motors distributed along the cell body. Through a combination of fluorescence imaging, force microscopy, and genetic manipulation, we show that membrane-bound cytoplasmic complexes consisting of motor and regulatory proteins are directionally transported down the axis of a cell at constant velocity. This intracellular motion is transmitted to the exterior of the cell and converted to traction forces on the substrate. Thus, this study demonstrates the existence of a conserved class of processive intracellular motors in bacteria and shows how these motors have been adapted to produce cell motility.
Asunto(s)
Adhesiones Focales/metabolismo , Locomoción/fisiología , Modelos Biológicos , Proteínas Motoras Moleculares/metabolismo , Myxococcus xanthus/fisiología , Adenosina Trifosfato/metabolismo , Western Blotting , Electroporación , Fluoresceínas , Fluorescencia , Concentración de Iones de Hidrógeno , Inmunoprecipitación , Quimografía , Microesferas , Plásmidos/genética , Transporte de Proteínas/fisiologíaRESUMEN
Retinal arteriovenous nicking (AVN) manifests as a reduced venular caliber of an arteriovenous crossing. AVNs are signs of many systemic, particularly cardiovascular diseases. Studies have shown that people with AVN are twice as likely to have a stroke. However, AVN classification faces two challenges. One is the lack of data, especially AVNs compared to the normal arteriovenous (AV) crossings. The other is the significant intra-class variations and minute inter-class differences. AVNs may look different in shape, scale, pose, and color. On the other hand, the AVN could be different from the normal AV crossing only by slight thinning of the vein. To address these challenges, first, we develop a data synthesis method to generate AV crossings, including normal and AVNs. Second, to mitigate the domain shift between the synthetic and real data, an edge-guided unsupervised domain adaptation network is designed to guide the transfer of domain invariant information. Third, a semantic contrastive learning branch (SCLB) is introduced and a set of semantically related images, as a semantic triplet, are input to the network simultaneously to guide the network to focus on the subtle differences in venular width and to ignore the differences in appearance. These strategies effectively mitigate the lack of data, domain shift between synthetic and real data, and significant intra- but minute inter-class differences. Extensive experiments have been performed to demonstrate the outstanding performance of the proposed method.
Asunto(s)
Enfermedades Cardiovasculares , Enfermedades de la Retina , Vena Retiniana , HumanosRESUMEN
AIM: To investigate the morphological characteristics of retinal vessels in patients with different severity of diabetic retinopathy (DR) and in patients with or without diabetic macular edema (DME). METHODS: The 239 eyes of DR patients and 100 eyes of healthy individuals were recruited for the study. The severity of DR patients was graded as mild, moderate and severe non-proliferative diabetic retinopathy (NPDR) according to the international clinical diabetic retinopathy (ICDR) disease severity scale classification, and retinal vascular morphology was quantitatively analyzed in ultra-wide field images using RU-net and transfer learning methods. The presence of DME was determined by optical coherence tomography (OCT), and differences in vascular morphological characteristics were compared between patients with and without DME. RESULTS: Retinal vessel segmentation using RU-net and transfer learning system had an accuracy of 99% and a Dice metric of 0.76. Compared with the healthy group, the DR group had smaller vessel angles (33.68±3.01 vs 37.78±1.60), smaller fractal dimension (Df) values (1.33±0.05 vs 1.41±0.03), less vessel density (1.12±0.44 vs 2.09±0.36) and fewer vascular branches (206.1±88.8 vs 396.5±91.3), all P<0.001. As the severity of DR increased, Df values decreased, P=0.031. No significant difference between the DME and non-DME groups were observed in vascular morphological characteristics. CONCLUSION: In this study, an artificial intelligence retinal vessel segmentation system is used with 99% accuracy, thus providing with relatively satisfactory performance in the evaluation of quantitative vascular morphology. DR patients have a tendency of vascular occlusion and dropout. The presence of DME does not compromise the integral retinal vascular pattern.
RESUMEN
Pressure ulcers present a significant human and economic challenge, lacking a reliable method for early detection. To address this, we developed a system capable of early detection by using cooling stimulation and dynamic data acquisition techniques to monitor blood perfusion and skin temperature. The system consists of laser speckle perfusion imaging and thermal imaging. And we performed simulations to demonstrate that the system is capable of detect tissue damage across multiple layers, from superficial to deep. Testing on a rabbit ear model demonstrated that this approach, which combines dynamic perfusion and temperature parameters, effectively distinguishes early pressure ulcer areas from normal skin with a significant p value of 0.0015. This distinction was more precise compared to methods relying solely on static parameters or one parameter. Our study thereby offers a promising advancement in the proactive management and prevention of pressure ulcers.
RESUMEN
Macular edema, a prevalent ocular complication observed in various retinal diseases, can lead to significant vision loss or blindness, necessitating accurate and timely diagnosis. Despite the potential of deep learning for segmentation of macular edema, challenges persist in accurately identifying lesion boundaries, especially in low-contrast and noisy regions, and in distinguishing between Inner Retinal Fluid (IRF), Sub-Retinal Fluid (SRF), and Pigment Epithelial Detachment (PED) lesions. To address these challenges, we present a novel approach, termed Semantic Uncertainty Guided Cross-Transformer Network (SuGCTNet), for the simultaneous segmentation of multi-class macular edema. Our proposed method comprises two key components, the semantic uncertainty guided attention module (SuGAM) and the Cross-Transformer module (CTM). The SuGAM module utilizes semantic uncertainty to allocate additional attention to regions with semantic ambiguity, improves the segmentation performance of these challenging areas. On the other hand, the CTM module capitalizes on both uncertainty information and multi-scale image features to enhance the overall continuity of the segmentation process, effectively minimizing feature confusion among different lesion types. Rigorous evaluation on public datasets and various OCT imaging device data demonstrates the superior performance of our proposed method compared to state-of-the-art approaches, highlighting its potential as a valuable tool for improving the accuracy and reproducibility of macular edema segmentation in clinical settings, and ultimately aiding in the early detection and diagnosis of macular edema-related diseases and associated retinal conditions.
Asunto(s)
Edema Macular , Tomografía de Coherencia Óptica , Humanos , Edema Macular/diagnóstico por imagen , Tomografía de Coherencia Óptica/métodos , Aprendizaje Profundo , Interpretación de Imagen Asistida por Computador/métodos , SemánticaRESUMEN
To investigate the circuit-level neural mechanisms of behavior, simultaneous imaging of neuronal activity in multiple cortical and subcortical regions is highly desired. Miniature head-mounted microscopes offer the capability of calcium imaging in freely behaving animals. However, implanting multiple microscopes on a mouse brain remains challenging due to space constraints and the cumbersome weight of the equipment. Here, we present TINIscope, a Tightly Integrated Neuronal Imaging microscope optimized for electronic and opto-mechanical design. With its compact and lightweight design of 0.43 g, TINIscope enables unprecedented simultaneous imaging of behavior-relevant activity in up to four brain regions in mice. Proof-of-concept experiments with TINIscope recorded over 1000 neurons in four hippocampal subregions and revealed concurrent activity patterns spanning across these regions. Moreover, we explored potential multi-modal experimental designs by integrating additional modules for optogenetics, electrical stimulation or local field potential recordings. Overall, TINIscope represents a timely and indispensable tool for studying the brain-wide interregional coordination that underlies unrestrained behaviors.
RESUMEN
Of the TRIM/RBCC family proteins taking part in a variety of cellular processes, TRIM50 is a stomach-specific member with no defined biological function. Our biochemical data demonstrated that TRIM50 is specifically expressed in gastric parietal cells and is predominantly localized in the tubulovesicular and canalicular membranes. In cultured cells ectopically expressing GFP-TRIM50, confocal microscopic imaging revealed dynamic movement of TRIM50-associated vesicles in a phosphoinositide 3-kinase-dependent manner. A protein overlay assay detected preferential binding of the PRY-SPRY domain from the TRIM50 C-terminal region to phosphatidylinositol species, suggesting that TRIM50 is involved in vesicular dynamics by sensing the phosphorylated state of phosphoinositol lipids. Trim50 knock-out mice retained normal histology in the gastric mucosa but exhibited impaired secretion of gastric acid. In response to histamine, Trim50 knock-out parietal cells generated deranged canaliculi, swollen microvilli lacking actin filaments, and excess multilamellar membrane complexes. Therefore, TRIM50 seems to play an essential role in tubulovesicular dynamics, promoting the formation of sophisticated canaliculi and microvilli during acid secretion in parietal cells.
Asunto(s)
Ácidos/química , Mucosa Gástrica/metabolismo , Regulación de la Expresión Génica , Proteínas de la Membrana/genética , Células Parietales Gástricas/citología , Animales , Linfocitos/citología , Proteínas de la Membrana/metabolismo , Ratones , Ratones Endogámicos C57BL , Ratones Noqueados , Microvellosidades/metabolismo , Modelos Biológicos , Fosfatidilinositol 3-Quinasas/metabolismo , Fosfatidilinositoles/química , Estructura Terciaria de Proteína , Transporte de Proteínas , Ratas , Ratas WistarRESUMEN
Purpose: The purpose of this study was to investigate the quantitative retinal vascular morphological characteristics of Retinopathy of Prematurity (ROP) and Familial Exudative Vitreoretinopathy (FEVR) in the newborn by the application of a deep learning network with artificial intelligence. Methods: Standard 130-degree fundus photographs centered on the optic disc were taken in the newborns. The deep learning network provided segmentation of the retinal vessels and the optic disc (OD). Based on the vessel segmentation, the vascular morphological characteristics, including avascular area, vessel angle, vessel density, fractal dimension (FD), and tortuosity, were automatically evaluated. Results: 201 eyes of FEVR, 289 eyes of ROP, and 195 eyes of healthy individuals were included in this study. The deep learning system of blood vessel segmentation had a sensitivity of 72% and a specificity of 99%. The vessel angle in the FEVR group was significantly smaller than that in the normal group and ROP group (37.43 ± 5.43 vs. 39.40 ± 5.61, 39.50 ± 5.58, P = 0.001, < 0.001 respectively). The normal group had the lowest vessel density, the ROP group was in between, and the FEVR group had the highest (2.64 ± 0.85, 2.97 ± 0.92, 3.37 ± 0.88 respectively). The FD was smaller in controls than in the FEVR and ROP groups (0.984 ± 0.039, 1.018 ± 0.039 and 1.016 ± 0.044 respectively, P < 0.001). The ROP group had the most tortuous vessels, while the FEVR group had the stiffest vessels, the controls were in the middle (11.61 ± 3.17, 8.37 ± 2.33 and 7.72 ± 1.57 respectively, P < 0.001). Conclusions: The deep learning technology used in this study has good performance in the quantitative analysis of vascular morphological characteristics in fundus photography. Vascular morphology was different in the newborns of FEVR and ROP compared to healthy individuals, which showed great clinical value for the differential diagnosis of ROP and FEVR.
RESUMEN
BACKGROUND: The purpose of this article is to develop a deep learning automatic segmentation model for the segmentation of Crohn's disease (CD) lesions in computed tomography enterography (CTE) images. Additionally, the radiomics features extracted from the segmented CD lesions will be analyzed and multiple machine learning classifiers will be built to distinguish CD activity. METHODS: This was a retrospective study with 2 sets of CTE image data. Segmentation datasets were used to establish nnU-Net neural network's automatic segmentation model. The classification dataset was processed using the automatic segmentation model to obtain segmentation results and extract radiomics features. The most optimal features were then selected to build 5 machine learning classifiers to distinguish CD activity. The performance of the automatic segmentation model was evaluated using the Dice similarity coefficient, while the performance of the machine learning classifier was evaluated using the area under the curve, sensitivity, specificity, and accuracy. RESULTS: The segmentation dataset had 84 CTE examinations of CD patients (mean age 31 ±â 13 yearsâ , 60 males), and the classification dataset had 193 (mean age 31 ±â 12 yearsâ , 136 males). The deep learning segmentation model achieved a Dice similarity coefficient of 0.824 on the testing set. The logistic regression model showed the best performance among the 5 classifiers in the testing set, with an area under the curve, sensitivity, specificity, and accuracy of 0.862, 0.697, 0.840, and 0.759, respectively. CONCLUSION: The automated segmentation model accurately segments CD lesions, and machine learning classifier distinguishes CD activity well. This method can assist radiologists in promptly and precisely evaluating CD activity.
The automatic segmentation and radiomics of computed tomography enterography images can assist radiologists in accurately and quickly identifying Crohn's disease lesions and evaluating Crohn's disease activity.
RESUMEN
Background: Recent advances in artificial intelligence and digital image processing have inspired the use of deep neural networks for segmentation tasks in multimodal medical imaging. Unlike natural images, multimodal medical images contain much richer information regarding different modal properties and therefore present more challenges for semantic segmentation. However, there is no report on systematic research that integrates multi-scaled and structured analysis of single-modal and multimodal medical images. Methods: We propose a deep neural network, named as Modality Preserving U-Net (MPU-Net), for modality-preserving analysis and segmentation of medical targets from multimodal medical images. The proposed MPU-Net consists of a modality preservation encoder (MPE) module that preserves the feature independency among the modalities and a modality fusion decoder (MFD) module that performs a multiscale feature fusion analysis for each modality in order to provide a rich feature representation for the final task. The effectiveness of such a single-modal preservation and multimodal fusion feature extraction approach is verified by multimodal segmentation experiments and an ablation study using brain tumor and prostate datasets from Medical Segmentation Decathlon (MSD). Results: The segmentation experiments demonstrated the superiority of MPU-Net over other methods in the segmentation tasks for multimodal medical images. In the brain tumor segmentation tasks, the Dice scores (DSCs) for the whole tumor (WT), the tumor core (TC) and the enhancing tumor (ET) regions were 89.42%, 86.92%, and 84.59%, respectively. In the meanwhile, the 95% Hausdorff distance (HD95) results were 3.530, 4.899 and 2.555, respectively. In the prostate segmentation tasks, the DSCs for the peripheral zone (PZ) and the transitional zone (TZ) of the prostate were 71.20% and 90.38%, respectively. In the meanwhile, the 95% HD95 results were 6.367 and 4.766, respectively. The ablation study showed that the combination of single-modal preservation and multimodal fusion methods improved the performance of multimodal medical image feature analysis. Conclusions: In the segmentation tasks using brain tumor and prostate datasets, the MPU-Net method has achieved the improved performance in comparison with the conventional methods, indicating its potential application for other segmentation tasks in multimodal medical images.
RESUMEN
Purpose: The purpose of this study is to investigate the retinal vascular morphological characteristics in high myopia patients of different severity. Methods: 317 eyes of high myopia patients and 104 eyes of healthy control subjects were included in this study. The severity of high myopia patients is classified into C0-C4 according to the Meta Analysis of the Pathologic Myopia (META-PM) classification and their vascular morphological characteristics in ultra-wide field imaging were analyzed using transfer learning methods and RU-net. Correlation with axial length (AL), best corrected visual acuity (BCVA) and age was analyzed. In addition, the vascular morphological characteristics of myopic choroidal neovascularization (mCNV) patients and their matched high myopia patients were compared. Results: The RU-net and transfer learning system of blood vessel segmentation had an accuracy of 98.24%, a sensitivity of 71.42%, a specificity of 99.37%, a precision of 73.68% and a F1 score of 72.29. Compared with healthy control group, high myopia group had smaller vessel angle (31.12 ± 2.27 vs. 32.33 ± 2.14), smaller fractal dimension (Df) (1.383 ± 0.060 vs. 1.424 ± 0.038), smaller vessel density (2.57 ± 0.96 vs. 3.92 ± 0.93) and fewer vascular branches (201.87 ± 75.92 vs. 271.31 ± 67.37), all P < 0.001. With the increase of myopia maculopathy severity, vessel angle, Df, vessel density and vascular branches significantly decreased (all P < 0.001). There were significant correlations of these characteristics with AL, BCVA and age. Patients with mCNV tended to have larger vessel density (P < 0.001) and more vascular branches (P = 0.045). Conclusion: The RU-net and transfer learning technology used in this study has an accuracy of 98.24%, thus has good performance in quantitative analysis of vascular morphological characteristics in Ultra-wide field images. Along with the increase of myopic maculopathy severity and the elongation of eyeball, vessel angle, Df, vessel density and vascular branches decreased. Myopic CNV patients have larger vessel density and more vascular branches.