A5004205811- Medical Image Segmentation Techniques
- Retinal Imaging and Analysis
- Cutaneous Melanoma Detection and Management
- AI in cancer detection
- Wound Healing and Treatments
- Radiomics and Machine Learning in Medical Imaging
- Color Science and Applications
- Retinal Diseases and Treatments
- Retinal and Optic Conditions
- Image Retrieval and Classification Techniques
- melanin and skin pigmentation
- CCD and CMOS Imaging Sensors
- Advanced Memory and Neural Computing
- Medical Imaging and Analysis
- Image Enhancement Techniques
- Neural dynamics and brain function
- Advanced Data Compression Techniques
- Industrial Vision Systems and Defect Detection
- Pressure Ulcer Prevention and Management
- Digital Imaging for Blood Diseases
- Smart Agriculture and AI
- Medical Imaging Techniques and Applications
- 3D Shape Modeling and Analysis
- Image and Signal Denoising Methods
- Infrared Thermography in Medicine
Universidad de Sevilla
2012-2022
University Frères Mentouri Constantine 1
2014
Ayuntamiento de Sevilla
2012
Instituto de Microelectrónica de Sevilla
2008
Event-driven visual sensors have attracted interest from a number of different research communities. They provide information in quite way conventional video systems consisting sequences still images rendered at given "frame rate." vision take inspiration biology. Each pixel sends out an event (spike) when it senses something meaningful is happening, without any notion frame. A special type event-driven sensor the so-called dynamic (DVS) where each computes relative changes light or...
In this paper different model-based methods of classification global patterns in dermoscopic images are proposed. Global identification is included the pattern analysis framework, melanoma diagnosis method most used among dermatologists. The modeling performed two senses: first a image modeled by finite symmetric conditional Markov model applied to L∗a∗b∗ color space and estimated parameters treated as features. turn, distribution these features supposed that follow models along lesion:...
This article investigates the suitability of local intensity distributions to analyze six emphysema classes in 342 CT scans obtained from 16 sites hosting scanners by 3 vendors and a total 9 specific models subjects with Chronic Obstructive Pulmonary Disease (COPD). We propose using kernel density estimation deal inherent sparsity histograms scarcely populated regions interest. validate our approach leave-one-subject-out classification experiments full-lung analyses. compare results recently...
In this paper a psychophysical experiment and multidimensional scaling (MDS) analysis are undergone to determine the physical characteristics that physicians employ diagnose burn depth. Subsequently, these translated into mathematical features, correlated with analysis. Finally, study verify ability of features classify burns is performed. study, space axes MDS has been developed. 74 images have represented in k-nearest neighbor classifier used images. A success rate 66.2% was obtained when...
. Skin cancer is the most common worldwide. One of non-melanoma tumors basal cell carcinoma (BCC), which accounts for 75% all skin cancers. There are many benign lesions that can be confused with these types cancers, leading to unnecessary biopsies. In this paper, a new method identify different BCC dermoscopic patterns present in lesion presented. addition, information applied classify into and non-BCC.
In this paper, a burn color image segmentation and classification system is proposed. The aim of the to separate wounds from healthy skin, distinguish among different types burns (burn depths). Digital photographs are used as inputs system. based on texture information, since these characteristics observed by physicians in order form diagnosis. A perceptually uniform space (L*u*v*) was used, Euclidean distances calculated correspond perceptual differences. After segmented, set features that...
Address-event representation (AER) is an emergent hardware technology which shows a high potential for providing in the near future solid technological substrate emulating brain-like processing structures. When used vision, AER sensors and processors are not restricted to capturing still image frames, as commercial frame-based video technology, but sense process visual information pixel-level event-based frameless manner. As result, vision practically simultaneous sensing, since there no...
Abstract Background The diagnosis of neuromuscular diseases is strongly based on the histological characterization muscle biopsies. However, this morphological analysis mostly a subjective process and difficult to quantify. We have tested if network science can provide novel framework extract useful information from biopsies, developing method that analyzes samples in an objective, automated, fast precise manner. Methods Our database consisted 102 biopsy images 70 individuals (including...
Diagnosis of neuromuscular diseases is based on subjective visual assessment biopsies from patients by the pathologist specialist. A system for objective analysis and classification muscular dystrophies neurogenic atrophies through muscle biopsy images fluorescence microscopy presented. The procedure starts with an accurate segmentation fibers using mathematical morphology a watershed transform. feature extraction step carried out in two parts: 24 features that pathologists take into account...
Color has great diagnostic significance in dermatoscopy. Several diagnosis methods are based on the colors detected within a lesion. Malignant lesions frequently show more than three colors, whereas benign lesions, or fewer usually observed. Black, red, white, and blue-gray found melanomas nevi. In this paper, method to automatically identify of lesion is presented. A color label identification problem proposed solved by maximizing posterior probability pixel belong label, given its value...
In this paper we propose the first bio-inspired six layer convolutional network (ConvNet) non-frame based that can be implemented with already physically available spike-based electronic devices. The system was designed to recognize people in three different positions: standing, lying or up-side down. inputs were spikes obtained a motion retina chip. We provide simulation results showing recognition delays of 16 milliseconds from stimulus onset (time-to-first spike) rate 94%. weight sharing...