Melatonin, as a plant growth regulator, is involved in stress resistance. We studied the effects of different concentrations (0, 10, 50, 100, 150, and 200 µmol · L−1) melatonin on physiological characteristics strawberry under cadmium (Cd) stress. The results represented that seedlings was inhibited Cd stress, seedling biomass, chlorophyll content activities antioxidant enzymes such superoxide dismutase (SOD), peroxidase (POD), catalase (CAT), ascorbate (APX) decreased. These toxic were,...

Multimodal retinal imaging plays an important role in ophthalmology. We propose a content-adaptive multimodal image registration method this paper that focuses on the globally coarse alignment and includes three weakly supervised neural networks for vessel segmentation, feature detection description, outlier rejection. apply proposed framework to register color fundus images with infrared reflectance fluorescein angiography images, compare it several conventional deep learning methods. Our...

In plants, anthocyanins and proanthocyanidins (PAs) play important roles in plant resistance to abiotic stress. this study, ozone (O3) treatments caused the up-regulation of Malus crabapple structural genes McANS, McCHI, McANR McF3H, which promoted anthocyanin PA accumulation. We identified WRKY transcription factor (TF) McWRKY71 by screening differentially expressed (DEGs) that were highly response O3 stress from an RNA sequencing (RNA-seq) analysis. Overexpressing increased 'Orin' apple...

Abstract The dysregulation of long noncoding RNAs (lncRNAs) and microRNAs (miRNAs) participates in the remodeling electrophysiological/ion channel cardiomyocytes during arrhythmia. lncRNA metastasis‐associated lung adenocarcinoma transcript 1 (MALAT1) is reported to be highly expressed myocardial ischemia‐reperfusion injury offsets cardioprotective effects fentanyl. However, roles MALAT1 its related miRNAs arrhythmia are poorly understood. In this study, overexpression was firstly indicated...

In biology and medicine research, detection identification of cancer cells plays an essential role to further analysis cell properties developing new drugs experiments. However, owing the adhesion among great changes in morphology, it is a very challenging task detect locate accurately, especially for area. this work, deep detector based on framework Faster R-CNN proposed, this, Circle Scanning Algorithm (CSA) presented redetection cells. And then series experiments are achieved. The results...

Multi-modal retinal image registration plays an important role in the ophthalmological diagnosis process. The conventional methods lack robustness aligning multi-modal images of various imaging qualities. Deep-learning have not been widely developed for this task, especially coarse-to-fine pipeline. To handle we propose a two-step method based on deep convolutional networks, including coarse alignment step and fine step. In step, global matrix is estimated by three sequentially connected...

In this paper, we first propose a quantitative indicator to measure the amount of prior information contained in wrapped phase map. Then, Edge-Enhanced Self-Attention Network is proposed for two-dimensional unwrapping. EESANet adopts symmetrical en-decoder architecture and uses self-designed Serried Residual Blocks as its basic block. We add Atrous Spatial Pyramid Pooling Positional network obtain long-distance dependency unwrapping, further Block enhance effective edge features addition,...

Summary The anthocyanin content is an important indicator of the nutritional value most fruits, including apple ( Malus domestica ). Anthocyanin synthesis coordinately regulated by light and various phytohormones. In this study on apple, we revealed antagonistic relationship between brassinosteroid (BR) signaling pathways, which mediated BRASSINAZOLE‐RESISTANT 1 (MdBZR1) B‐box protein MdCOL6. exogenous application brassinolide inhibited high‐light‐induced accumulation in red‐fleshed...

Multimodal image registration plays an important role in diagnosing and treating ophthalmologic diseases. In this paper, a deep learning framework for multimodal retinal is proposed. The consists of segmentation network, feature detection description outlier rejection which focuses only on the globally coarse alignment step using perspective transformation. We apply proposed to register color fundus images with infrared reflectance compare it state-of-the-art conventional learning-based...

In multi-modal retinal image registration task, there are two major challenges, i.e., poor performance in finding correspondence due to inconsistent features, and lack of labeled data for training learning-based models. this paper, we propose a joint vessel segmentation deformable model based on CNN built under the framework weakly supervised style transfer learning perceptual loss. segmentation, loss guides generate maps that look authentic, helps transform images different modalities into...

The ability to accurately overlay one modality retinal image another is critical in ophthalmology. Our previous framework achieved the state-of-the-art results for multimodal registration. However, it requires human-annotated labels due supervised approach of work. In this paper, we propose a self-supervised retina registration method alleviate burdens time and expense prepare training data, that is, aiming automatically register images without any human annotations. Specially, focus on...

Foveal Avascular Zone (FAZ) is a crucial indicator for retinal disease detection and accurate automatic FAZ segmentation has significant impact in clinical applications. Apart from the binary map, vessel map can provide further information. To simultaneously implement segmentation, an end-to-end trained network proposed to achieve unsupervised supervised segmentation. Due lack of labels, style transfer with consistency loss Then achieved U-Net structure based on Two superficial layer OCTA...

Given a child's and couple's facial photos, tri-subject kinship verification aims to determine the existence of blood relation between child couple. Different from existing methods which model inheritance process among three persons in separate stages only use simple features, this work establishes inspired by genetics measure similarity one step. Meanwhile, high-dimensional features are incorporated into seek for better performance. Experiment results demonstrate effectiveness our approach.

Visual kinship recognition aims to identify blood relatives from facial images. Its practical application-- like in law-enforcement, video surveillance, automatic family album management, and more-- has motivated many researchers put forth effort on the topic as of recent. In this paper, we focus a new view visual technology: kin-based face generation. Specifically, propose two-stage kin-face generation model predict appearance child given pair parents. The first stage includes deep...

Purpose: The purpose of this study was to evaluate the ability align two types retinal images taken on different platforms; color fundus (CF) photographs and infrared scanning laser ophthalmoscope (IR SLO) using mathematical warping artificial intelligence (AI). Methods: We collected 109 matched pairs CF IR SLO images. An AI algorithm utilizing separate networks developed. A style transfer network (STN) used segment vessel structures. registration segmented each. Neither a ground truth...

Road detection from images is a challenging task in computer vision. Previous methods are not robust, because their features and classifiers cannot adapt to different circumstances. To overcome this problem, we propose apply unsupervised feature learning for road detection. Specifically, develop an improved encoding function add selection process obtain robust discriminative features. Besides, segmentation algorithm proposed extract regions the learned maps, which tree structure established...

Integrating deep learning with traditional machine methods is an intriguing research direction. For example, PCANet and LDANet adopts Principal Component Analysis (PCA) Fisher Linear Discriminant (LDA) to learn convolutional kernels separately. It not reasonable adopt LDA filter in each layer, local features of images from different classes may be similar, such as background areas. Therefore, it meaningful only when all the patches carry information whole image. However, our knowledge, there...

Abstract Purpose This study aimed to compare a new Artificial Intelligence (AI) method conventional mathematical warping in accurately overlaying peripheral retinal vessels from two different imaging devices: confocal scanning laser ophthalmoscope (cSLO) wide-field images and SLO ultra-wide field images. Methods Images were captured using the Heidelberg Spectralis 55-degree field-of-view Optos field. The was performed Random Sample Consensus—Sample Consensus sets (RANSAC-SC). compared an AI...