A5046393559- Radiomics and Machine Learning in Medical Imaging
- COVID-19 diagnosis using AI
- Advanced Neuroimaging Techniques and Applications
- Lung Cancer Diagnosis and Treatment
- Advanced Neural Network Applications
- Brain Tumor Detection and Classification
- Medical Image Segmentation Techniques
- Medical Imaging Techniques and Applications
- Advanced Image and Video Retrieval Techniques
- Advanced MRI Techniques and Applications
- Reservoir Engineering and Simulation Methods
- Natural Language Processing Techniques
- Topic Modeling
- Functional Brain Connectivity Studies
- Fetal and Pediatric Neurological Disorders
- Image and Signal Denoising Methods
- Optical measurement and interference techniques
- Domain Adaptation and Few-Shot Learning
- Indoor and Outdoor Localization Technologies
- Remote-Sensing Image Classification
- Optical Polarization and Ellipsometry
- Advanced Image Processing Techniques
- Multimodal Machine Learning Applications
- Data Quality and Management
- Oil and Gas Production Techniques
Universidade Estadual de Campinas (UNICAMP)
2019-2025
University of Iowa
2023
University of Trás-os-Montes and Alto Douro
2017
Universidade Federal da Bahia
2017
Methodist University of Piracicaba
2015
Abstract The field of supervised automated medical imaging segmentation suffers from relatively small datasets with ground truth labels. This is especially true for challenging problems that target structures low contrast and ambiguous boundaries, such as glass opacities consolidation in chest computed tomography images. In this work, we make available the first public dataset opacity lungs Long COVID patients. Iowa-UNICAMP (LongCIU) was built by three independent expert annotators, blindly...
In natural language processing (NLP), there is a need for more resources in Portuguese, since much of the data used state-of-the-art research other languages. this paper, we pretrain T5 model on BrWac corpus, an extensive collection web pages and evaluate its performance against Portuguese pretrained models multilingual three different tasks. We show that our have significantly better over original models. Moreover, demonstrate positive impact using vocabulary. Our code are available at...
Background: Hippocampus segmentation on magnetic resonance imaging is of key importance for the diagnosis, treatment decision and investigation neuropsychiatric disorders. Automatic an active research field, with many recent models using deep learning. Most current state-of-the art hippocampus methods train their healthy or Alzheimer's disease patients from public datasets. This raises question whether these are capable recognizing a different domain, that epilepsy resection.New Method: In...
In this letter, we introduce a novel method for visual simultaneous localization and mapping (SLAM)-so-called Air-SSLAM-which exploits stereo camera configuration. contrast to monocular SLAM, scale definition 3-D information are issues that can be more easily dealt with in cameras. Air-SSLAM starts from computing keypoints the correspondent descriptors over pair of images, using good features-to-track rotated-binary robust-independent elementary features, respectively. Then map is created by...
The COVID-19 pandemic generated research interest in automated models to perform classification and segmentation from medical imaging of patients, However, applications real-world scenarios are still needed. We describe the development deployment decision support system. A partnership with a Brazilian radiologist consortium, gave us access 1000s labeled computed tomography (CT) X-ray images São Paulo Hospitals. system used EfficientNet EfficientDet networks, state-of-the-art convolutional...
Corpus callosum (CC) segmentation is an important first step of MRI-based analysis, however most available automated methods and tools perform its on the midsagittal slice only. Additionally, few volumetric CC work T1-weighted images, what requires additional registering T1 mask over diffusion tensor images (DTI) when conducting any DTI-based analysis. This presents a method corpus using modified U-Net data, such as Fractional Anisotropy (FA), Mean Difusivity (MD) Mode (MO). The model was...
Identifying and characterizing brain fiber bundles can help to understand many diseases conditions. An important step in this process is the estimation of orientations using Diffusion-Weighted Magnetic Resonance Imaging (DW-MRI). However, obtaining robust orientation estimates demands high-resolution data, leading lengthy acquisitions that are not always clinically available. In work, we explore use automated angular super resolution from faster overcome challenge. Using publicly available...
The fusion of multispectral (MSI) and hyperspectral (HSI) images is a crucial technique in various fields such as remote sensing, medical imaging, agricultural monitoring. MSI captures light across several specific spectral bands, while HSI provides detailed information contiguous bands. Combining these two types leverages the high spatial resolution rich content HSI, creating single, high-resolution image that both spatially spectrally informative. Traditional wavelet based methods often...
Hippocampus segmentation plays a key role in diagnosing various brain disorders such as Alzheimer's disease, epilepsy, multiple sclerosis, cancer, depression and others. Nowadays, is still mainly performed manually by specialists. Segmentation done experts considered to be gold-standard when evaluating automated methods, buts it time consuming arduos task, requiring specialized personnel. In recent years, efforts have been made achieve reliable segmentation. For years the best performing...
The thalamus is a subcortical brain structure linked to the motor system. Since certain changes within this are related diseases, such as multiple sclerosis and Parkinson’s, characterization of thalamus—e.g., shape assessment—is crucial step in relevant studies applications, including medical research surgical planning. A robust reliable thalamus-segmentation method therefore, required meet these demands. Despite presenting low contrast for particular structure, T1-weighted imaging still...
The COVID-19 pandemic response highlighted the potential of deep learning methods in facilitating diagnosis, prognosis and understanding lung diseases through automated segmentation pulmonary structures lesions chest computed tomography (CT). Automated separation lesion into ground-glass opacity (GGO) consolidation is hindered due to labor-intensive subjective nature this task, resulting scarce availability ground truth for supervised learning. To tackle problem, we propose MEDPSeg. MEDPSeg...