RESUMEN
The tear fluid reservoir (TFR) under the sclera lens is a unique characteristic providing optical neutralization of any aberrations from corneal irregularities. Anterior segment optical coherence tomography (AS-OCT) has become an important imaging modality for sclera lens fitting and visual rehabilitation therapy in both optometry and ophthalmology. Herein, we aimed to investigate whether deep learning can be used to segment the TFR from healthy and keratoconus eyes, with irregular corneal surfaces, in OCT images. Using AS-OCT, a dataset of 31850 images from 52 healthy and 46 keratoconus eyes, during sclera lens wear, was obtained and labeled with our previously developed algorithm of semi-automatic segmentation. A custom-improved U-shape network architecture with a full-range multi-scale feature-enhanced module (FMFE-Unet) was designed and trained. A hybrid loss function was designed to focus training on the TFR, to tackle the class imbalance problem. The experiments on our database showed an IoU, precision, specificity, and recall of 0.9426, 0.9678, 0.9965, and 0.9731, respectively. Furthermore, FMFE-Unet was found to outperform the other two state-of-the-art methods and ablation models, suggesting its strength in segmenting the TFR under the sclera lens depicted on OCT images. The application of deep learning for TFR segmentation in OCT images provides a powerful tool to assess changes in the dynamic tear film under the sclera lens, improving the efficiency and accuracy of lens fitting, and thus supporting the promotion of sclera lenses in clinical practice.
RESUMEN
Background: The ciliary muscle plays a role in changing the shape of the crystalline lens to maintain the clear retinal image during near work. Studying the dynamic changes of the ciliary muscle during accommodation is necessary for understanding the mechanism of presbyopia. Optical coherence tomography (OCT) has been frequently used to image the ciliary muscle and its changes during accommodation in vivo. However, the segmentation process is cumbersome and time-consuming due to the large image data sets and the impact of low imaging quality. Objectives: This study aimed to establish a fully automatic method for segmenting and quantifying the ciliary muscle on the basis of optical coherence tomography (OCT) images. Design: A perspective cross-sectional study. Methods: In this study, 3500 signed images were used to develop a deep learning system. A novel deep learning algorithm was created from the widely used U-net and a full-resolution residual network to realize automatic segmentation and quantification of the ciliary muscle. Finally, the algorithm-predicted results and manual annotation were compared. Results: For segmentation performed by the system, the total mean pixel value difference (PVD) was 1.12, and the Dice coefficient, intersection over union (IoU), and sensitivity values were 93.8%, 88.7%, and 93.9%, respectively. The performance of the system was comparable with that of experienced specialists. The system could also successfully segment ciliary muscle images and quantify ciliary muscle thickness changes during accommodation. Conclusion: We developed an automatic segmentation framework for the ciliary muscle that can be used to analyze the morphological parameters of the ciliary muscle and its dynamic changes during accommodation.
RESUMEN
Purpose: To clarify the effects of acute hyperglycemia on the responses of choroidal structural components and vascularity index during light modulation in healthy participants using techniques including image binarization and artificial intelligence (AI) segmentation based on swept-source optical coherence tomography (SS-OCT). Methods: Twenty-four eyes of 24 healthy participants were imaged at different stages after ambient light, 40 min of dark adaptation, and 5 min of light adaptation in two imaging sessions: control and after receiving 75 g of oral glucose solution. The choroidal structural parameters, including luminal volume (LV), stromal volume (SV), total choroidal volume (TCV), and choroidal vascularity index (CVI) within a 6 mm area were determined using a custom algorithm based on image binarization and AI segmentation of SS-OCT. These measurements were compared among the conditions after adjusting for axial length, age to identify the differences. Results: In the dark, CVI decreased (-0.36 ± 0.09%) significantly in acute hyperglycemia compared to the control condition. During the transition to ambient light, there was an increasing trend in the choroidal parameters compared with the control experiment. However, only TCV (0.38 ± 0.17 mm3) and LV (0.27 ± 0.10 mm3) showed a significant increase at the time point of 5 min after ambient light. Conclusion: Analysis of choroidal structural parameters and CVI based on SS-OCT images is a potentially powerful method to objectively reflect subtle changes in neurovascular coupling between the choroid and photoreceptor during dark adaptation.
Asunto(s)
Hiperglucemia , Tomografía de Coherencia Óptica , Humanos , Tomografía de Coherencia Óptica/métodos , Inteligencia Artificial , Coroides/diagnóstico por imagen , Coroides/irrigación sanguínea , Enfermedad Aguda , Adaptación Ocular , Hiperglucemia/diagnóstico por imagenRESUMEN
PURPOSE: To investigate the impact of penetration and image analysis in different optical coherence tomography (OCT) instruments on the measurement of choroidal vascularity parameters. METHODS: Twenty-three healthy volunteers were imaged using two swept-source OCTs and one spectral-domain OCT. A fully automatic segmentation method based on ResNet-UNet and Niblack local threshold binarization was performed to quantify the relevant choroidal vascular parameters, including choroidal vascularity index, total choroidal volume, and luminal volume. The intraclass correlation coefficient (ICC) and coefficient of repeatability (COR) were used to analyze the repeatability and consistency of automatic and manual segmentation, respectively. RESULTS: Both swept-source OCT devices showed good consistency of luminal volume and total choroidal volume measurements (all ICC value >0.98 with COR% < 8.53%) based on manual segmentation, whereas the consistency of the spectral-domain OCT was lower (ICC value <0.60 with COR% > 40%), which was greatly improved after using the automatic algorithm (ICC value >0.99 with COR% < 4%). The repeatability of choroidal vascularity index obtained from different OCT images using manual or automatic segmentation showed good agreement (all ICC values >0.85), whereas the choroidal vascularity index measurement from the spectral-domain OCT was larger than the other two swept-source OCT devices (ICC value <0.65). CONCLUSION: For healthy youngsters, the penetration of OCT plays a role in the measurement precision for choroidal vascularity parameters, and automatic segmentation can improve the ability of choroidal boundary identification with deficient penetration, suggesting these factors need to be considered in clinical work.
Asunto(s)
Coroides , Tomografía de Coherencia Óptica , Algoritmos , Coroides/irrigación sanguínea , Humanos , Procesamiento de Imagen Asistido por Computador , Reproducibilidad de los Resultados , Tomografía de Coherencia Óptica/métodosRESUMEN
1,6-bis(trichlorosilyl)hexane (C6Cl), polystyrene (PS), and cross-linked polystyrene (CPS) were investigated as gate dielectric modified layers for high performance organic transistors. The influence of the surface energy, roughness and morphology on the charge transport of the organic thin-film transistors (OTFTs) was investigated. The surface energy and roughness both affect the grain size of the pentacene films which will control the charge carrier mobility of the devices. Pentacene thin-film transistors fabricated on the CPS modified dielectric layers exhibited charge carrier mobility as high as 1.11 cm2 V-1 s-1. The bias stress stability for the CPS devices shows that the drain current only decays 1% after 1530 s and the mobility never decreases until 13530 s.
Asunto(s)
Microelectrodos , Nanopartículas/química , Naftacenos/química , Poliestirenos/química , Transistores Electrónicos , Reactivos de Enlaces Cruzados/química , Impedancia Eléctrica , Diseño de Equipo , Análisis de Falla de Equipo , Ensayo de Materiales , Nanopartículas/ultraestructura , Tamaño de la PartículaRESUMEN
OBJECTIVE: To study the effect of Limax lyophilized powder on bronchial asthma. METHODS: The allergic asthma model was established in guinea pigs by combined utilization of aluminum hydroxide and egg albumin to investigate the effect of Limax lyophilized powder on the bronchial flow and on the level of inflammator in bronchoalveolar lavage and serum. RESULTS: The mortality of asthma laboratory guinea pigs was reduced and the incubation period of asthma was extended significantly in Limax lyophilized powder groups. Its antiasthmatic effect was as efficient as the control drug (aminophylline). The leucocyte count was decreased in peripheral blood and the bronchoalveolar lavage fluid. The infiltration of pulmonary tissues eosinophil was also significantly reduced. Further more,the most efficient effects was showed in Limax lyophilized powder at the moderate dosage (63 mg/kg). The bronchial perfusion flow was increased and the level of IL-2 and IL-4 in blood serum and bronchoalveolar lavage fluid was decreased obviously in the aminophylline group and Limax lyophilized powder groups at moderate and high dosage. CONCLUSION: Limax lyophilized powder could inhibit bronchial asthma evidently.