Stefan Udristoiu

Automation of the medical diagnosis process using semantic image interpretation

This paper is a part of a complex study of developing methods for semantic interpretation of medical images, to permit the semi-automatic diagnosis. The first objective of the study is to develop new methods for medical image segmentation and a set of visual features. The second objective consists of developing a unifying framework for semantic images annotation, to be used in the process of medical diagnosis. The developed diagnosis method is based on on semantic pattern rules capable to discover associations between visual features of medical images and their diagnoses. Although we present the results achieved in endoscopic images analysis, our methods can be used to analyze other types of medical images. The prototype system was applied to real datasets and the results show high accuracy.

Mapping Image Low-Level Descriptors to Semantic Concepts

Our goal is to organize the image contents semantically. In this paper, we propose a method to classify the images semantically, using the C-fuzzy algorithm to segment the natural scenes into perceptually uniform regions. The low-level characteristics that are taken into account are: color, texture, shape, absolute spatial arrangement, spatial coherency, and dimension. Since humans are the ultimate users of most image retrieval systems, it is important to organize the contents semantically, according to meaningful categories. This requires an understanding of the important semantic categories that humans use for image classification, and the extraction of meaningful image features that can discriminate between these categories. A lot of experiments, in which the human subjects had to group images into semantic categories and to explain the criteria for their choice, were realized. From these experiments, we identify the semantic categories (landscapes, animals, flowers, etc), the semantic indicators or intermediate descriptors and their visual characteristics.

Image Annotation by Learning Rules from Regions Patterns

The modeling of multimedia and especially the semantic gap between the visual features and semantic concepts become an important domain due to the quantity of visual digital content, which speedily grows. In this paper, the analysis and semantic annotation of images are studied. The development of methods for colour image annotations based on learning represents the main contribution of the paper. The developed algorithms generate semantic pattern rules that identify high-level image concepts. A semantic pattern rule is a combination of images’ region patterns that identifies semantic concepts. Our methods are not limited to any specific domain and they can be applied in any field.

Improving an image retrieval system by integrating semantic features

To develop image navigation systems, we need tools to realize the semantic relationship between user and database. In this paper, it is presented an indexing schema of images and a simple semantic vocabulary that permits to the user to introduce the cognitive dimension in the retrieval process. A lot of experiments, in which the human subjects had to group images into semantic categories and to explain the criteria for their choice, were realized. From these experiments, we identify the semantic categories (landscapes, animals, flowers, etc), the semantic indicators or intermediate descriptors and their visual characteristics.

Predicting Students’ Results Using Rough Sets Theory

This paper proposes the utilization of rough set theory for predicting student scholar performance. The rough set theory is a powerful approach that permits the searching for patterns in e-learning database using the minimal length principles. Searching for models with small size is performed by means of many different kinds of reducts that generate the decision rules capable for identifying the final student grade.

Rule-Based Methods for the Computer Assisted Diagnosis of Medical Images

In this paper we study a method for automatic diagnosis based on edoscopies of medical images. We develop algorithms that automatically generate rules that identify medical diagnosis. A semantic rule is a combination of semantic indicator values that identifies the image diagnosis. These rules are represented in Prolog and can be shared and modified depending on the updates in a respective domain. The experiments are realized on colour endoscopies from digestive apparatus, recording promising results.

Discovering Interesting Patterns in an e-Learning System

This paper presents a method for discovering interesting patterns in an e-learning system using a clustering method based on variable precision rough set theory and association rules. The information from each cluster is then used to generate interesting rules in order to help teachers in the learning process and to understand students’ behavior. To accomplish this, a database with students enrolled for a “Database” course is analyzed and the presented method discovers rules of the students’ behavior regarding the assignments, course quizzes, and also the rules of student’s interaction with teacher and other students.