Abstract The wood-based furniture manufacturing industries prioritize quality of production to meet higher market demands. Identifying various types edge-glued wooden panel defects are a challenge for human worker or camera. Several studies have shown that the detection with low, high, normal, overlong, short is identified but residue and bluntness highly challenging. Thus, present model identifies by computer vision and/or deep learning, whereas learning based decide pass having better...

Open-box models in medical domain have high acceptance and demand by many examiners. Even though the accuracy predicted most of convolutional neural network (CNN) is high, it still not convincing as detail discussion regarding outcome semi-transparent functioning process. As pneumonia one top contagious infection that makes population affected due to low immunity. Therefore, goal this paper implement an interpretable classification using eXplainable AI (XAI-ICP). Thus, XAI-ICP highly...

Street lighting is a fundamental aspect of security systems in homes, industrial facilities, and public places. To detect parking lot occupancy outdoor environments, street light control plays crucial role smart surveillance applications that can perform robustly extreme environments. However, traditional are mostly implemented for environments using costly sensor-based techniques. This study uses the Jetson TX2 to develop method accurately identify streetlights assist detection, thereby...

Abstract Background Agitation and sedation management is critical in intensive care as it affects patient safety. Traditional nursing assessments suffer from low frequency subjectivity. Automating these can boost unit (ICU) efficiency, treatment capacity, Objectives The aim of this study was to develop a machine-learning based assessment agitation sedation. Methods Using data the Taichung Veterans General Hospital ICU database (2020), an ensemble learning model developed for classifying...

Objective The aim of this study was to develop an artificial intelligence–based model detect the presence acute respiratory distress syndrome (ARDS) using clinical data and chest X-ray (CXR) data. Method transfer learning method used train a convolutional neural network (CNN) with external image dataset extract features. Then, last layer fine-tuned determine probability ARDS. were trained three machine algorithms—eXtreme Gradient Boosting (XGB), random forest (RF), logistic regression...

This study aimed to develop an early prediction model for identifying patients with bloodstream infections. The data resource was taken from 2015 2019 at Taichung Veterans General Hospital, and a total of 1647 infection episodes 3552 non-bloodstream in the intensive care unit (ICU) were included development evaluation. During analysis, 30 clinical variables selected, including patients’ basic characteristics, vital signs, laboratory data, information. Five machine learning algorithms applied...

Abstract Objective To address the challenge of assessing sedation status in critically ill patients intensive care unit (ICU), we aimed to develop a non-contact automatic classifier agitation using artificial intelligence and deep learning. Methods We collected video recordings ICU cut them into 30-second (30-s) 2-second (2-s) segments. All segments were annotated with as “Attention” “Non-attention”. After transforming movement quantification, constructed models classifiers Threshold, Random...

Early detection of rapidly progressive kidney disease is key to improving the renal outcome and reducing complications in adult patients with type 2 diabetes mellitus (T2DM). We aimed construct a 6-month machine learning (ML) predictive model for risk need nephrology referral T2DM an initial estimated glomerular filtration rate (eGFR) ≥ 60 mL/min/1.73 m2. extracted medical features from electronic records (EMR), cohort was divided into training/validation testing data set develop validate...

Evaluating several vital signs and chest X-ray (CXR) reports regularly to determine the recovery of pneumonia patients at general wards is a challenge for doctors. A recent study shows identification by history symptoms including signs, CXR, other clinical parameters, but they lack predicting status after starting treatment. The goal this paper provide prediction system early affected patient's discharge from hospital within 7 days or late more than days. This aims design multimodal data...

Video surveillance systems are deployed at many places such as airports, train stations, and malls for security monitoring purposes. However, it is laborious to search retrieve persons in multicamera systems, especially with cluttered backgrounds appearance variations among multiple cameras. To solve these problems, this paper proposes a person retrieval method that extracts the attributes of masked image using an instance segmentation module each object interest. It uses color type clothes...

Background: Antinuclear antibody pattern recognition is vital for autoimmune disease diagnosis but labor-intensive manual interpretation. To develop an automated system, we established machine learning models based on the International Consensus Antibody Patterns (ICAP) at a competent level, mixed patterns recognition, and evaluated their consistency with human reading. Methods: 51,694 epithelial cells (HEp-2) cell images assigned by experienced medical technologists collected in center were...

As the globe undergoes extreme climate changes, disaster events and their scale continue to increase; therefore, it has become imperative devise prevention measures.Traditional flood monitoring devices, while operating in harsh environments, are often influenced by changes weather conditions such light, rain, fog.Consequently, recorded images blurred or damaged, which increases possibility of errors judgment delays hazard mitigation process.In this study, an automated identification method...

Sheet metal-based manufacturing industries operate on several varieties of sheet metal parts. Previously QR codes stickers were put parts for identification by manual workers as per their respective shape and size features, thus ensuring synchronized raw material flow the process. However, identifying a particular type part based its different features is still challenge trained operator. Currently, in market, there exist some automation solutions solving such kind problem but are incapable...

There is strong demand for real-time suspicious tracking across multiple cameras in intelligent video surveillance public areas, such as universities, airports and factories. Most criminal events show that the nature of behavior are carried out by un-known people who try to hide themselves much possible. Previous learning-based studies collected a large volume data set train learning model detect humans but failed recognize newcomers. also several feature-based aimed identify within-camera...

This paper focuses on flood-region detection using monitoring images. However, adverse weather affects the outcome of image segmentation methods. In this paper, we present an experimental comparison outdoor visual sensing system region-growing methods with two different growing rules-namely, GrowCut and RegGro. For each rule, several tests lens-stained scenes were performed, taking into account analyzing conditions system. The influence was analyzed, highlighting their effect rules....

Early prediction of clinical deterioration such as adverse events (AEs), improves patient safety. National Warning Score (NEWS) is widely used to predict AEs based on the aggregation 6 physiological parameters. We took same parameters features for AE using deep learning algorithms (AEP-DLA) among hospitalized adult patients. The aim this study get better performance than traditional naïve mathematical calculations by introducing novel vital sign data preprocessing schemes. retrospectively...

A distinct security protocol is necessary for the exponential growth in intelligent edge devices. In particular, autonomous devices need to address significant concern function smoothly high market demand. Nevertheless, increase connected has made cloud networks more complex and suffer from information processing delay. Therefore, goal of this work design a novel server-less mutual authentication networks. The aim demonstrate an amongst smart within solution addresses applications cars,...

Surveillance systems are everywhere deployed for criminal prevention and suspect tracking specific security events. The recognition of the target person cross multiple cameras remains a challenge because large spatio-temporal uncertainty, which means entry exit persons in unpredictable. Therefore, this study design implement software system that help users to continuous stable track same across cameras. Firstly, leverages YOLO Correlation filter suspicious single camera. features also...

Sheet metal-based products serve as a major portion of the furniture market and maintain higher quality standards by being competitive. During industrial processes, while converting sheet metal to an end product, new defects are observed thus need be identified carefully. Recent studies have shown scratches, bumps, pollution/dust identified, but orange peel present overall challenge. So our model identifies dust using computer vision algorithms, whereas defect detection with deep learning...

The Journal of the National Science Foundation Sri Lanka publishes results research in all aspects and Technology. journal also has a website at http://www.nsf.gov.lk/. 2021 Impact Factor: 0.682The JNSF provides immediate open access to its content on principle that making freely available public supports greater global exchange knowledge.Cover :Leatherback (a), green (b - adult & f hatchling), hawksbill (c) olive ridley (d) turtles who nest Lankan beaches, turtle crawl marks (e) by-catch...

This paper proposes a framework for an Image-based Flood Alarm (IFA) that includes bi-seeded region-based image segmentation the extraction of water region interest from image, as well alarm classifier identifying degree flood risk. When risk reaches predetermined threshold, response message reports to main EWS end-user decision support.

Severe weather conditions greatly impair the performance of outdoor imaging. In this study, two region-based image segmentation methods, Grow Cut and Region Growing (RegGro), were applied to rain scenes. This study demonstrates that accuracy depends on fog stains. severe rainfall periods, heavy reduced overall quality, both methods yielded failure. The results show are effective for segmenting objects in images captured under poor conditions. Both have unique advantages disadvantages stain...