A5062735262- Digital Radiography and Breast Imaging
- Medical Imaging Techniques and Applications
- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Advanced MRI Techniques and Applications
- MRI in cancer diagnosis
- Radiation Dose and Imaging
- Breast Cancer Treatment Studies
- Advanced X-ray and CT Imaging
- Pediatric Urology and Nephrology Studies
- Innovative Educational Technologies
- Kidney Stones and Urolithiasis Treatments
- Lanthanide and Transition Metal Complexes
- Educational Games and Gamification
- Advanced Image Fusion Techniques
- Image and Signal Denoising Methods
University of Lisbon
2015-2025
University College London
2015
The objective of this study is to propose an advanced image enhancement strategy address the challenge reducing radiation doses in pediatric renal scintigraphy. Data from a public dynamic scintigraphy database were used. Based on noisier images, four denoising neural networks (DnCNN, UDnCNN, DUDnCNN, and AttnGAN) evaluated. To evaluate quality noise reduction, with minimal detail loss, kidney signal-to-noise ratio (SNR) multiscale structural similarity (MS-SSIM) Although all reduced noise,...
Breast cancer remains a leading cause of mortality among women, with molecular subtypes significantly influencing prognosis and treatment strategies. Currently, identifying the subtype requires biopsy—a specialized, expensive, time-consuming procedure, often yielding to results that must be supported additional biopsies due technique errors or tumor heterogeneity. This study introduces novel approach for predicting breast using mammography images advanced artificial intelligence (AI)...
Microcalcification clusters (MCs) are among the most important biomarkers for breast cancer, especially in cases of nonpalpable lesions. The vast majority deep learning studies on digital tomosynthesis (DBT) focused detecting and classifying lesions, soft-tissue small regions interest previously selected. Only about 25% specific to MCs, all them based classification preselected regions. Classifying whole image according presence or absence MCs is a difficult task due size information present...
Purpose: Compressed sensing (CS) is a new approach in medical imaging which allows sparse image to be reconstructed from undersampled data. Total variation (TV) based minimization algorithms are the one CS technique that has achieved great success due its virtue of preserving edges while reducing noise. The purpose this work implement and evaluate performance TV filter able increase signal difference noise ratio (SDNR) digital breast tomosynthesis (DBT) images. Methods: Assuming Poisson...
Breast cancer is the most commonly diagnosed worldwide. The therapy used and its success depend highly on histology of tumor. This study aimed to explore potential predicting molecular subtype breast using radiomic features extracted from screening digital mammography (DM) images. A retrospective was performed OPTIMAM Mammography Image Database (OMI-DB). Four binary classification tasks were performed: luminal vs. non-luminal A, B B, TNBC non-TNBC, HER2 non-HER2. Feature selection carried...
Currently, breast cancer is the most commonly diagnosed type of worldwide. Digital Breast Tomosynthesis (DBT) has been widely accepted as a stand-alone modality to replace Mammography, particularly in denser breasts. However, image quality improvement provided by DBT accompanied an increase radiation dose for patient. Here, method based on 2D Total Variation (2D TV) minimization improve without need was proposed. Two phantoms were used acquire data at different ranges (0.88-2.19 mGy Gammex...
Digital Breast Tomosynthesis (DBT) presents out-of-plane artifacts caused by features of high intensity. Given observed data and knowledge about the point spread function (PSF), deconvolution techniques recover from a blurred version. However, correct PSF is difficult to achieve these methods amplify noise. When no information available PSF, blind can be used. Additionally, Total Variation (TV) minimization algorithms have achieved great success due its virtue preserving edges while reducing...
Digital Breast Tomosynthesis (DBT) is a developing imaging modality which produces 3D images of breast. Iterative image reconstruction techniques, such as Algebraic technique (ART), have been proposed to help increasing success in detecting masses and micro-calcifications. To enhance the quality reconstructed image, total variation (TV) minimization was applied by ART. Nowadays, number published papers dealing with TV on ART (ART+TV <sub xmlns:mml="http://www.w3.org/1998/Math/MathML"...
The Partial Volume (PV) effect in Positron Emission Tomography (PET) imaging leads to loss quantification accuracy, which manifests PV effects (small objects occupy partially the sensitive volume of instrument, resulting blurred images). Simultaneous acquisition PET and Magnetic Resonance Imaging (MRI) produces concurrent metabolic anatomical information. latter has proved be very helpful for correction effects. Currently, there are several techniques used correction. They can applied...
Slice by slice visualization of Digital Breast Tomosynthesis (DBT) data is time consuming and can hamper the interpretation lesions such as clusters microcalcifications. With a object through multiple angles, 3D volume rendering (VR) provides an intuitive understanding underlying at once. VR may play important complementary role in breast cancer diagnosis. Transfer functions (TFs) are critical parameter finding good TFs major challenge. The purpose this work to study methodology...
3D volume rendering may represent a complementary option in the visualization of Digital Breast Tomosynthesis (DBT) examinations by providing an understanding underlying data at once. Rendering parameters directly influence quality rendered images. The purpose this work is to study two these (voxel dimension z direction and sampling distance) on DBT data. Both were studied with real phantom one clinical set. voxel size was changed from 0.085 × 1.0 mm3 using ten interpolation functions...