A5103266934- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- 3D Shape Modeling and Analysis
- Image Retrieval and Classification Techniques
- Underwater Acoustics Research
- Underwater Vehicles and Communication Systems
- Medical Imaging and Analysis
- Medical Imaging Techniques and Applications
- Machine Learning and Algorithms
- Image Processing and 3D Reconstruction
- Smart Agriculture and AI
- Environmental Education and Sustainability
- Textile materials and evaluations
- AI in cancer detection
- Dental Radiography and Imaging
- Color perception and design
- Gene expression and cancer classification
- Career Development and Diversity
- Single-cell and spatial transcriptomics
- Innovative Human-Technology Interaction
- Chemotherapy-induced cardiotoxicity and mitigation
- Face and Expression Recognition
- PARP inhibition in cancer therapy
- Image and Signal Denoising Methods
- Robotics and Sensor-Based Localization
Universitat de València
2011-2024
Duke University
2004-2006
Science, Technology, Engineering, and Mathematics (STEM) are key disciplines towards tackling the challenges related to Sustainable Development Goals. However, evidence shows that women enrolling in these a smaller percentage than men, especially Engineering fields. As stated by United Nations Women section, increasing number of studying working STEM fields is fundamental achieving better solutions global challenges, since potential for innovation larger. In this paper, we present Girls4STEM...
A data-adaptive algorithm is presented for the selection of basis functions and training data used in classifier design with application to sensing mine-like targets a side-scan sonar. Automatic detection using sonar imagery complicated by variability target, clutter, background signatures. Specifically, strong dependence on environmental conditions vitiates assumption that one may perform priori separate collected previously. In this paper, novel active-learning developed based kernel...
Mine-like object classification from sidescan sonar images is of great interest for mine counter measure (MCM) operations. Because the shadow cast by an often most distinct feature a image, standard procedure to perform based on features extracted shadow. The can then be performed extracting and comparing this training data determine object. In paper, superellipse fitting approach classifying mine-like objects in presented. Superellipses provide compact efficient way representing different...
This article shares the strategy for mainstreaming Sustainable Development Goals (SDGs) at University of Valencia (UV), which, although limited in its scale, may compel other Higher Education Institutions to think technological and social progress aligned with 2030 Agenda. It explicates a process driven by UV, on occasion 75th anniversary United Nations (UN), collaboration Service Geospatial, Information, Telecommunications Technologies from UN Support Base (Spain) prepare online event: “The...
This paper proposes one possibility to generalize the morphological operations (particularly, dilation, erosion, opening, and closing) color images. First, properties of a desirable generalization are stated brief review is done on former approaches. Then, method explained, which based total ordering colors in an image induced by its histogram; this valid for just may present problems smoothly coloured To solve these drawbacks refinement consisting smoothing histogram using joint several...
Anatomical atlases are 3D volumes or shapes representing an organ structure of the human body. They contain either prototypical shape object interest together with other its statistical variations (statistical atlas) a probability map belonging to (probabilistic atlas). Probabilistic mostly built simple estimations only involving data at each spatial location. A new method for probabilistic atlas construction that uses generalized linear model is proposed. This aims improve estimation be...
This works deals with the concept of liver segmentation by using a priori information based on probabilistic atlases and learning previous steps. A atlas is here understood as probability or membership map that tells how likely point belongs to shape drawn from distribution at hand. We devise procedure segment Perfusion Magnetic Resonance images combines both: algorithm global simpler steps, local close segmented slices finally mathematical morphology procedure, namely viscous...
The development of automatic and reliable algorithms for the detection segmentation vertebrae are great importance prior to any diagnostic task. However, an important problem found accurately segment is presence ribs in thoracic region. To overcome this problem, a probabilistic atlas spine has been developed dealing with proximity other structures, special focus on suppression.The data sets used consist Computed Tomography images corresponding 21 patients suffering from spinal metastases....
Numerous studies have been published which, separately, investigate the influence of molecular features on oncological and cardiac pathologies. Nevertheless, relationship between both families diseases at level is an emerging area within onco-cardiology/cardio-oncology. This paper presents a new open-source database that aims to organize curated information concerning validated in patients involved cancer cardiovascular diseases. Entities like gene, variation, drug, study others are modelled...
Partitioning around medoids (PAM) is one of the most widely used and successful clustering method in many fields. One its key advantages that it only requires a distance or dissimilarity between individuals, fact cluster centers are actual points data set means they can be taken as reliable representatives their classes. However, wider application hampered by large amount memory needed to store matrix (quadratic on number individuals) also high computational cost computing such and, less...
Brain lesions are life threatening diseases. Traditional diagnosis of brain is performed visually by neuro-radiologists. Nowadays, advanced technologies and the progress in magnetic resonance imaging provide computer aided using automated methods that can detect segment abnormal regions from different medical images. Among several techniques, machine learning based flexible efficient. Therefore, this paper, we present a review on techniques applied for detection segmentation images with...
Spine is a structure commonly involved in several prevalent diseases. In clinical diagnosis, therapy, and surgical intervention, the identification segmentation of vertebral bodies are crucial steps. However, automatic detailed vertebrae challenging task, especially due to proximity corresponding ribs other structures such as blood vessels. this study, overcome these problems, probabilistic atlas spine, including cervical, thoracic lumbar has been built introduce anatomical knowledge...
Functional Magnetic Resonance (fMR) is a medical image technique in which contrast injected the vascular system so that blood diffusion along it can be observed as variations of signal intensity. The uptake agent are used early detection tumorous tissue. For diagnostic to accurate, successive volumes must correctly registered. binary registration prior segmentation 3D fMR data required. Here we present local level-set method preserves details and edges, with its multi-scale version has...
Most contemporary web applications are primarily coded in interpreted languages (JavaScript, PHP, Python...) and initiated by the server. This requires solving persistence issue: HTTP/HTTPS is a stateless protocol but user identity computational state across consecutive requests must be preserved, typically using cookies and/or backend database servers.This work develops embedded applications: single compiled executable programs that encapsulate These language (in our case, C++). They...
This study contributes to the Health 4.0 paradigm by enhancing precision of cell nuclei detection in histopathological images, a critical step digital pathology. The presented approach is characterized combination deep learning with traditional analytic classifiers. Traditional methods histopathology rely heavily on manual inspection expert histopathologists. While has revolutionized this process offering rapid and accurate detections, its black-box nature often results lack...