Accurate classification of Alzheimer's disease (AD) and its prodromal stage, mild cognitive impairment (MCI), plays a critical role in possibly preventing progression memory improving quality life for AD patients. Among many research tasks, it is particular interest to identify noninvasive imaging biomarkers diagnosis. In this paper, we present robust deep learning system different stages patients based on MRI PET scans. We utilized the dropout technique improve classical by weight...

We present a multi-robot system for GPS-denied search and rescue under the forest canopy. Forests are particularly challenging environments collaborative exploration mapping, in large part due to existence of severe perceptual aliasing which hinders reliable loop closure detection mutual localization map fusion. Our proposed features unmanned aerial vehicles (UAVs) that perform onboard sensing, estimation, planning. When communication is available, each UAV transmits compressed tree-based...

Optical Coherence Tomography (OCT) is a noninvasive test that takes cross-section pictures of the retina layer eye and allows ophthalmologists to diagnose based on retina's layers. Therefore, it an important modality for detection quantification retinal diseases abnormalities. Since OCT provides several images each patient, time consuming work analyze images. This paper proposes deep learning models categorize patients' into four categories such as Choroidal neovascularization (CNV),...

The growth of Internet Things (IoT) brings the promise a wide range new recommender systems due to expected 57 billion smart connected devices by 2025. In this paper, we propose IoT platform for supporting real-time system. To illustrate effectiveness our proposed platform, present prototype implementation and tourism application demonstrate entire process from user event data collection notification/recommendations provision. We conducted several experiments including notification system...

Glaucoma is one of the most common eye diseases that can cause irreversible vision loss due to damage optic nerve. Ophthalmologists consider a cup disc ratio greater than 0.3 be suggestive glaucoma. Unfortunately, there high variability among ophthalmologists in estimating since it not easy reliably measure and areas fundus image. Therefore, this paper proposes automatic methods segment areas. There are two steps estimate ratio: region interest (ROI) area detection (where center) from image,...

Glaucoma is a leading cause of irreversible vision loss that gradually damages the optic nerve. In ophthalmic fundus images, measurements cup to disc (CD) ratio, CD area neuroretinal rim (RD) and thickness are key measures screen for potential glaucomatous damage. We propose an automatic method using deep learning algorithms segment estimate measures. The proposed comprises three steps: Region Interest (ROI) (location disc) detection from image Mask R-CNN, segmentation ROI Multiscale Average...

Age-related macular degeneration (AMD) and Diabetic edema (DME) are retinal disease that can cause permanent vision loss. AMD is the leading of irreversible loss in individuals aged 65 above DME largest caused visual patients with diabetes world. Early detection treatment important to treat or delay progress. Ophthalmologists use Optical Coherence Tomography (OCT) as one key modalities diagnose diseases decide whether perform anti-VEGF therapy since it provides cross-section patients' retina...

Protein function prediction is the important problem in modern biology. In this paper, un-normalized, symmetric normalized, and random walk graph Laplacian based semi-supervised learning methods will be applied to integrated network combined from multiple networks predict functions of all yeast proteins these networks. These are created Pfam domain structure, co-participation a protein complex, protein-protein interaction network, genetic cell cycle gene expression measurements. Multiple...

Most network-based protein (or gene) function prediction methods are based on the assumption that labels of two adjacent proteins in network likely to be same. However, assuming pairwise relationship between or genes is not complete. The information a group show very similar patterns expression and tend have functions (i.e. functional modules) missed. natural way overcoming loss above represent gene data as hypergraph. Thus, this paper, three un-normalized, random walk, symmetric normalized...

Optical Coherence Tomography (OCT) is a non-invasive imaging technique that uses light waves to capture cross-sectional images of patients' retina layers, allowing for the diagnosis various retinal diseases. Ophthalmologists use OCT decide whether perform anti-Vascular Endothelial Growth Factor therapy. However, it time-consuming work analyze since provides several each patient. This paper proposes an ensemble learning (EL) model, based on three deep models, categorize into four categories...

We all know that VGG deep neural network is one of the most advanced and powerful learning models popular used in computer vision. However, cost training serving sometimes considerable due to large sets parameters. Therefore, practice, it necessary provide constructive methods compress these models, while keeping same level accuracy. In this paper, we study on use SVD CUR decomposition techniques compare them with original networks image classification problems. Experimental results,...

This paper presents a hybrid technique of combining the BERT embedding method and graph convolutional neural network. combination is then employed to solve text classification problem. Initially, we apply whole corpus in order transform all texts into numerical vectors. Then, network will be applied these vectors classify their appropriate classes. Especially, our approach, need only few labeled for model training. For illustration, this paper, use BBC news IMDB movie reviews datasets...

Most network-based machine learning methods are based on the assumption that labels of two adjacent vertices in network likely to be same.However, assuming pairwise relationship between is not complete.The information a group show very similar patterns and tend have missed.The natural way overcoming loss above represent given data as hypergraph.However, representing dataset hypergraph will lead perfection.The number hyper-edges may large; hence this high time complexity clustering or...

f autonomous systems using trust and trustworthiness is the focus of Autonomy Teaming TRAjectories for Complex Trusted Operational Reliability (ATTRACTOR), a new NASA Convergent Aeronautical Solutions (CAS) Project. One critical research element ATTRACTOR explainability decision-making across relevant subsystems an system. The ability to explain why system makes decision needed establish basis safely complete mission. Convolutional Neural Networks (CNNs) are popular visual object classifiers...

Ophthalmologists use the optic disc to cup ratio as one factor diagnose glaucoma. Optic in fundus images is area where blood vessels and nerve fibers enter retina. A (the diameter of divided by disc) greater than 0.3 considered be suggestive Therefore, we are developing automatic methods estimate areas, ratio. There four steps ratio: region interest (ROI) detection (where center) from image, segmentation ROI, area, estimation. This paper proposes an automated method segment ROI using deep...

Most network-based protein (or gene) function prediction methods are based on the assumption that labels of two adjacent proteins in network likely to be same. However, assuming pairwise relationship between or genes is not complete, information a group show very similar patterns expression and tend have functions (i.e. functional modules) missed. The natural way overcoming loss above represent gene data as hypergraph. Thus, this paper, three un-normalized, random walk, symmetric normalized...

Glaucoma is a chronic retinal disease that gradually damages the optic nerve. It leading cause of irreversible loss vision. In ophthalmic fundus images, cup to disc ratio measured around nerve key measure used screen for glaucomatous damages. Unfortunately, there high subjectivity among ophthalmologists in estimating this due challenges making reliable and measurements. To minimize this, we propose an automatic method using deep learning ensemble segment cup. The proposed comprises two...