Laboratory of Image Data Science (LIDS)

While the advances in data acquisition and storage have provided large image databases for research and development in many areas of the Sciences and Engineering, effective machine learning often requires a large number of labeled examples (e.g., image objects, superpixels, or subimages). This constitutes a crucial drawback, especially when specialists in the application domain are needed to interactively isolate and annotate those examples in each image.

We have investigated image processing, visual analytics, machine learning, and pattern recognition methods to minimize user interaction, without losing user control and understanding over the machine's actions, and, whenever it is possible, automate solutions of problems from other areas, such as Medicine, Biology, Geology, and Agriculture.

The current projects can be categorized in the following research topics.

Interactive Machine Learning

Supervised machine learning methods often dismiss human interaction during the learning process, compromising the understanding and confidence of the specialists in the machine's actions. We aim at considerably reducing the required number of labeled examples for machine learning to build explainable and reliable decision-making (-support) systems based on image analysis. We have investigated computational methods that exploit the superior...

Automated Diagnosis of Gastrintestinal Parasites

Gastrintestinal parasites in humans and animals constitute a serious problem, which can cause physical and mental deficiencies, diseases, and death. Their diagnosis currently rely on the visual analysis of optical microscopy slides, being very susceptible to human errors. We have investigated the full automation of the diagnosis of gastrintestinal parasites in human and animals. The solution for the problem...

3D Medical Image Segmentation and Analysis

The visualization and quantification of anatomical structures in 3D medical images is paramount to understand the natural course of the disease, plan a treatment, and study the effects of the treatment. This project investigates methods for image segmentation of anatomical structures of the brain and thorax. We are interested in establishing a standard of brain asymmetry from MR images...

Plant Image Analysis at Different Scales

Imaging technologies have provided ways to measure complex plant traits at different scales. Such measurements allow us to understand how plants grow and how this growth is affected by the environment conditions. We have investigated image analysis techniques to understand how plant development occurs from stem cells in the shoot apical meristem using confocal microscopy images, and to phenotype plant...

Graph-based Image Processing and Machine Learning

The interpretation of an image as a graph may occur at different levels where the graph nodes are pixels or superpixels. Even an image database may be interpreted as a graph whose nodes are the images or image objects. In any case, each node is represented by a set of image attributes and the arcs describe binary relations between...