Ocular imaging has developed rapidly and plays a critical role in clinical care ocular disease management. Development of image processing technologies pertinent to diseases paved the way for automated diagnostic systems including detection techniques using deep learning (DL) approaches. The prevalence an abnormal tissue layer conjunctiva, known as pterygium eye disease, is increasing due lack awareness. Despite non-cancerous/benign nature pterygium, diagnosis from ophthalmologist still...

<span>Segmentation of blood vessels (BVs) from retinal image is one the important steps in developing a computer-assisted diagnosis system and has been widely researched especially for implementing automatic BV segmentation methods. This paper proposes an improvement to existing (RBV) method by combining trainable B-COSFIRE filter with adaptive thresholding The proposed can automatically configure its selectivity given prototype pattern be detected. Its performance comparable many...

Traffic signs images when captured in real environment are small size compared to other objects. Thus, making it difficult be accurately detected, more so identified. Of recent, the convolutional neural network (CNN) has been tremendous progress object detection due its high accuracy and fast execution. YOLO (You Only Look Once) is an method, which uses CNN core module suited for of traffic sign environment. In this paper, we implemented YOLOv3 framework identification Malaysian With...

Seven quality control (7QC) tools are used for reducing defects during manufacturing. These highly effective in productivity and improvement. In this case, the study of 7QC was applied an automotive factory order to reduce paint shop defects. Within four months production line inspected, were categorized successfully applied, overall defect rate by 70%. Although every tool important, cause effect diagram responsible finding root causes

Recent technological advancements have changed significantly the way news is produced, consumed, and disseminated. Frequent on-spot reporting has been enabled, which smartphones can access anywhere anytime. News categorization or classification help in its proper timely dissemination. This study evaluates compares category predictors' performance based on four supervised machine learning models. We choose a standard dataset of British Broadcasting Corporation (BBC) consisting five...

Retinal image analysis has been widely used for early detection and diagnosis of multiple systemic diseases. Accurate vessel extraction in retinal is a crucial step towards fully automated system. This work affords an efficient unsupervised method extracting blood vessels from images by combining existing Gabor Wavelet (GW) with automatic thresholding. Green channel extracted color to produce feature using GW. Both green undergo vessel-enhancement order highlight vessels. Next, the two...

Body Sensor Networks (BSNs) are increasingly being used in pervasive sensing environments including healthcare, sports, wellbeing, and gaming. Activity segmentation using BSN is challenging the use of manual annotation subjective error prone. In this paper, we investigate a semi-supervised activity method Multiple Eigenspace (MES) technique based on Principal Components Analysis (PCA). Results show that can reliably perform classification results HMMs demonstrate practical value proposed technique.

For timely diagnosis of retinal disease, routine monitoring people with high risk should be put in place. To assist the ophthalmologists performing analysis efficiently and accurately, numerous studies have been conducted to propose an automated system. One crucial steps for such a system is accurate detection blood vessels from image. In this paper, we investigated use automatic binarization methods on pre-processed fundus image detect vessels. Three were study, namely Otsu’s method,...

There has been an increasing interest by local Malaysian towards beekeeping of Meliponine, which is a tribe stingless bee. When it comes to advice on beekeeping, these locals (mostly novice) depends inputs from only handful key experts aging). This paper presents part project that aims develop visual-based expert system, may help in tasks. One the first processes such system image segmentation. Being small size, segmentation Meliponine its natural surrounding difficult be accurately...

The prevalence of chronic kidney disease (CKD) is rising in Malaysia. Early detection necessary to prevent progression, especially terms cardiovascular (CV) risk, the main cause death end-stage renal (ESRD). Retinal changes have proven be a good predictor CKD whereas cardiac biomarkers are useful risk stratification. We aimed demonstrate correlation between retinal and with CKD.This single-centre cross-sectional study was conducted among patients stages 3, 4, 5 (not on dialysis) from...

This study presents an interesting approach to identifying the Genus of Stingless bee aided by machine learning technology. The conventional way or "Lebah Kelulut" relied on face-to-face meetings with local experts. particular process is considered be outdated and time consuming. Thus, proposed solution incorporated tool called "TensorFlow Object Detection API". provided Google TensorFlow uses Faster Region-based Convolutional Neural Network (Faster R-CNN), which incorporates Region Proposal...

Ocular diseases are becoming more common these days, especially the ones directly related to diabetes such as diabetic retinopathy (DR). This can be attributed steady increase in number of cases worldwide. DR treated if diagnosed early enough prevent further damages retina which could lead total blindness left untreated. Automatic segmentation retinal blood vessels from fundus image is useful towards development an efficient and accurate computer-assisted diagnosis system. In this study, we...

<ns3:p><ns3:bold>Background</ns3:bold><ns3:bold>: </ns3:bold>Retinal images can be considered as one of the reliable indicators for symptoms many ocular diseases such diabetic retinopathy, macular degeneration and glaucoma. By analysing tracking changes important structures on a retinal image, detected in timely manner which helps physicians plan early treatment better disease control. One landmarks image is optic disc (OD), must localised to estimate vessel parameters width tortuosity. This...

The need for computerized insect identification and classification system primarily with stingless bees is gaining attention. To determine the genus of a bee or “lebah kelulut” in Malay language has traditionally been by face-to-face consultation experts. Such task relies on manual that inefficient time consuming. This paper looks into via image processing method. proposed solution templates thorax automatically identified cuticle tessellation Lepidotrigona more than 80% accuracy 8 different...

Bee keeping has gained quite an attention by localMalaysian especially towards a particular bee species known as Meliponine, which is tribe of stingless bee.Up until now, the identification this still heavily relies on input handful key experts.Many studies have shown that honey contains beneficial nutrients can combat against various diseases, makes it very popular and valuable compared to normal honey.This study aims aid process developing visual-based guidance system.The main system will...

<ns4:p><ns4:bold>Background</ns4:bold><ns4:bold>:</ns4:bold><ns4:bold> </ns4:bold>By diagnosing using fundus images, ophthalmologists can possibly detect symptoms of retinal diseases such as diabetic retinopathy, age-related macular degeneration, and detachment. A number studies have also found some links between image analysis data other underlying systemic cardiovascular diseases, including hypertension kidney dysfunction. Now that imaging technology is advancing further, more cameras are...