OPINION article Front. Pharmacol., 12 June 2020Sec. Experimental Pharmacology and Drug Discovery Volume 11 - 2020 | https://doi.org/10.3389/fphar.2020.00912

To detect quickly and accurately “Yuluxiang” pear fruits in non-structural environments, a lightweight YOLO-GEW detection model is proposed to address issues such as similar fruit color leaves, bagging, complex environments. This improves upon YOLOv8s by using GhostNet its backbone for extracting features of the pears. Additionally, an EMA attention mechanism was added before fusing each feature neck section make focus more on target information fruits, thereby improving recognition ability...

This study introduces an improved YOLOv8 model YOLO-FSL for rapid detection of plum maturity under natural conditions. Firstly, the Faster Net Block module was introduced into C2f YOLOv8s backbone network, forming C2f_Faster module; Secondly, squeeze and excitation (SE) attention mechanism after each added in network; Finally, pruned using channel pruning algorithm to simplify ensure efficiency. The experimental results show that mean average precision (mAP) size is 91.2% 4.1 MB (Mega Byte),...

The intelligent detection of young peaches is the main technology fruit-thinning robots, which crucial for enhancing peach fruit quality and reducing labor costs. This study presents lightweight YOLO-PEM model based on YOLOv8s to achieve high-precision automatic “Okubo” peaches. Firstly, C2f_P module was devised by partial convolution (PConv), replacing all C2f modules in model’s lightweight. Secondly, embedding efficient multi-scale attention (EMA) C2f_P_1 backbone network enhanced feature...

Fraud, Waste, and Abuse (FWA) detection is a significant yet challenging problem in the health insurance industry. An essential step FWA to check whether medication clinically reasonable with respect diagnosis. Currently, human experts sufficient medical knowledge are required perform this task. To reduce cost, inspectors tend build an intelligent system detect suspicious claims inappropriate diagnoses/medications automatically.The aim of study was develop automated method for making use...

To assess the performance of models for early prediction acute kidney injury (AKI) in Intensive Care Unit (ICU) setting.Data were collected from Medical Information Mart (MIMIC)-III database all patients aged ≥18 years who had their serum creatinine (SCr) level measured 72 h following ICU admission. Those with existing conditions disease upon admission excluded our analyses. Seventeen predictor variables comprising patient demographics and physiological indicators selected on basis Kidney...

The efficient identification of the field flat jujube is first condition to realize its automated picking. Consequently, a lightweight algorithm target based on improved YOLOv5 (you only look once) proposed meet requirements high-accuracy and low-complexity. At first, method solves imbalance data distribution by improving methods enhancement. Then, improve accuracy model, we adjust structure number Concentrated-Comprehensive Convolution Block modules in backbone network, introduce attention...

A pepper quality detection and classification model based on transfer learning combined with convolutional neural network is proposed as a solution for low efficiency of manual sorting at the current stage. The dataset was amplified data pre-processing methods including rotation, luminance switch, contrast ratio switch. To improve training speed precision, optimized fine-tuned VGG 16 in this research, applied after parameter optimization, comparative analysis performed by combining ResNet50,...

Fruits with various maturity levels coexist among the harvested jujubes, and have different tastes uses. Manual grading has a low efficiency strong subjectivity. The number of “Hupingzao” jujubes between is unbalanced, which affects performance classifier. To solve above issue, class balance loss (CB) was used to improve MobileNet V2 network, transfer learning strategy train model. model optimized based on selection an optimizer rate. achieved best classification results using AdamW rate...

To perform accurate and synchronous detection of the soluble solid contents (SSC) in fresh jujubes at different stages maturity, hyperspectral imaging was used to establish robust models. The combined data constituting four maturation were build grid-search least squares support vector machine (GS-LS-SVM) model. determination coefficient (Rp2), root-mean-square error (RMSEP), residual predictive deviation (RPD) prediction set for samples overall 0.98, 1.10%, 7.85, respectively. Furthermore,...

In order to investigate the effects of concentration and temperature treatment carboxymethyl-chitosan (CM-C) solution on quality "Niuxin" persimmon during cold storage, fruits were soaked in CM-C different concentrations, study changes indicators including SSC, total titratable acid, firmness, soluble tannin, protopectin, peroxidase (POD), polyphenol oxidase (PPO), relative electrical conductivity storage. Results showed that heat treatment, combined heat, could effectively maintain storage...

Abstract Background Health information avoidance is common in real life, but because it not always conducive to health promotion and maintenance, people often actively switch acquisition. Understanding this process of active change can facilitate intervention unreasonable behaviors. However, studies so far have mostly focused on why how takes place, little known about the from We thus use a grounded theory approach (GT) explore generate theoretical framework capable illustrating stages...

Abstract Objective This study aimed to develop and validate a quantitative index system for evaluating the data quality of Electronic Medical Records (EMR) in disease risk prediction using Machine Learning (ML). Materials methods The was developed four steps: (1) preliminary outlined based on literature review; (2) we utilized Delphi method structure indicators at all levels; (3) weights these were determined Analytic Hierarchy Process (AHP) method; (4) empirically validated real-world EMR...

Glioblastoma multiforme (GBM) is the most aggressive adult brain tumor and presents significant treatment challenges due to its poor prognosis heterogeneity. Despite rapid development of deep learning, integrating multi-omics with whole slide images (WSIs) for survival prediction remains difficult. This study aims improve GBM by WSIs multi-omic data through incorporation biological pathway knowledge. Utilizing multiple instance learning co-attention mechanisms, we initiated integration...

Abstract In order to implement the real‐time detection of moisture content in fresh walnuts during storage, a visible–near infrared spectroscopy (NIR) kinetic model at room temperature was established. Eight characteristic wavelengths were optimized by two‐dimensional correlation technology, using as external perturbation. The average different storage times analyzed, and first‐order reaction established ( , RMSEP = 0.119 ). information index bands coupled establish partial least square...