A5003973442- Radiomics and Machine Learning in Medical Imaging
- Medical Imaging Techniques and Applications
- Advanced X-ray and CT Imaging
- AI in cancer detection
- Digital Radiography and Breast Imaging
- MRI in cancer diagnosis
- Advanced Neuroimaging Techniques and Applications
- Medical Image Segmentation Techniques
- Brain Metastases and Treatment
- Brain Tumor Detection and Classification
- Cerebral Palsy and Movement Disorders
- Anomaly Detection Techniques and Applications
- Advanced MRI Techniques and Applications
- Botulinum Toxin and Related Neurological Disorders
- Advanced Radiotherapy Techniques
- Head and Neck Cancer Studies
- Nuclear Physics and Applications
- Radiopharmaceutical Chemistry and Applications
- Electricity Theft Detection Techniques
- Medical Imaging and Analysis
- Global Cancer Incidence and Screening
- COVID-19 diagnosis using AI
- Imbalanced Data Classification Techniques
- Cardiac Imaging and Diagnostics
- Radiation Detection and Scintillator Technologies
Universitat Politècnica de València
2013-2024
Centro Tecnológico de Investigación, Desarrollo e Innovación en tecnologías de la Información y las Comunicaciones (TIC)
2022-2024
Instituto de Instrumentación para Imagen Molecular
2019
Consejo Superior de Investigaciones Científicas
2019
Universitat de València
2015-2018
Purpose To develop a classification model using texture features and support vector machine in contrast‐enhanced T1‐weighted images to differentiate between brain metastasis radiation necrosis. Methods Texture were extracted from 115 lesions: 32 of them previously diagnosed as necrosis, 23 radiation‐treated 60 untreated metastases; including total 179 derived six analysis methods. A feature selection technique based on was used obtain subset that provide optimal performance. Results The...
Purpose To investigate the ability of texture analysis to differentiate between infarcted nonviable, viable, and remote segments on cardiac cine magnetic resonance imaging ( MRI ). Methods This retrospective study included 50 patients suffering chronic myocardial infarction. The data were randomly split into training (30 patients) testing (20 sets. left ventricular myocardium was segmented according 17‐segment model in both late gadolinium enhancement LGE ) . Infarcted regions identified...
Brain metastases are occasionally detected before diagnosing their primary site of origin. In these cases, simple visual examination medical images the is not enough to identify cancer, so an extensive evaluation needed. To avoid this procedure, a radiomics approach on magnetic resonance (MR) metastatic lesions proposed classify two most frequent origins (lung cancer and melanoma). study, 50 T1-weighted MR brain from 30 patients were analyzed: 27 lung 23 melanoma A total 43 statistical...
Breast cancer is a major health concern worldwide. Mammography, cost-effective and accurate tool, crucial in combating this issue. However, low contrast, noise, artifacts can limit the diagnostic capabilities of radiologists. Computer-Aided Diagnosis (CAD) systems have been developed to overcome these challenges, with outlining breast being critical step for further analysis. This study introduces SAM-breast model, an adaptation Segment Anything Model (SAM) segmenting region mammograms....
Breast density assessed from digital mammograms is a biomarker for higher risk of developing breast cancer. Experienced radiologists assess using the Image and Data System (BI-RADS) categories. Supervised learning algorithms have been developed with this objective in mind, however, performance these depends on quality ground-truth information which usually labeled by expert readers. These labels are noisy approximations ground truth, as there often intra- inter-reader variability among...
Breast density assessed from digital mammograms is a known biomarker related to higher risk of developing breast cancer. Supervised learning algorithms have been implemented determine this. However, the performance these depends on quality ground-truth information, which expert readers usually provide. These labels are noisy approximations ground truth, as there both intra- and inter-observer variability among them. Thus, it crucial provide reliable method measure mammograms. This paper...
Detection of infarcted myocardium in the left ventricle is achieved with delayed enhancement magnetic resonance imaging (DE-MRI). However, manual segmentation tedious and prone to variability. We studied three texture analysis methods (run-length matrix, co-occurrence autoregressive model) combination histogram features characterize myocardium. evaluated 10 patients chronic infarction select most discriminative train a support vector machine (SVM) classifier. The classifier model was then...
Detection of brain metastases in patients with undiagnosed primary cancer is unusual but still an existing phenomenon. In these cases, identifying the site origin non-feasible by visual examination magnetic resonance (MR) images. Recently, radiomics has been proposed to analyze differences among classes visually imperceptible imaging characteristics. this study we analyzed 46 T1-weighted MR images from 29 patients: lung and 17 breast origin. A total 43 texture features were extracted...
In positron emission tomography (PET) image reconstruction, loss of contrast and incorrect quantification activity are produced due to the effect Compton scattering. Thus, scatter correction becomes essential improve quality. this work, a machine learning approach based on supervised has been considered for in simulated multi-ring PET system using cylindrical phantom. Using positional energy information from both photons detected as input data, we able label each coincidence True or...
Synthetic computed tomography (CT) images derived from magnetic resonance (MRI) are of interest for radiotherapy planning and positron emission (PET) attenuation correction. In recent years, deep learning implementations have demonstrated improvement over atlas-based segmentation-based methods. Nevertheless, several open questions remain to be addressed, such as which is the best MRI sequences neural network architectures. this work, we compared performance different combinations two common...
In the financial world, some transactions can be fraudulent. Appropriate detection for such these could help to reduce them. An illegal trade made through intrinsic features and by using structural information related way followed during its lifecycle. Fusing both heterogeneous sources contribute accuracy improvement. Nowadays, graph-based technologies allow fusion of geometric linked nodes transitions between nodes. The article aims validate new Deep Learning based on graphs Directed Graph...