Gabriel Mihai

Topic Map for Medical E-Learning

The paper presents original ways of using a modern concept - topic map - in medical e-learning. The topic map is mainly used for visualizing a thesaurus containing medical terms. The topic map is built and populated in an original manner, mapping an xml file that can be downloaded free, to an xtm file that contains the structure of the topic map. Only a part of the MeSH thesaurus was used, namely the part that includes the medical diagnosis’s names. The student can navigate through topic map depending on its interest subject, having in this way big advantages. The paper presents also how to use the topic map for semantic querying of a multimedia database with medical information and images. For retrieving the interest information this access path can be combined with another modern solution: the content-based visual query on the multimedia medical database. Combining these possibilities to access a database with medical data and images, allows students to see images and associated information in a simple and direct manner. The students are stimulated to learn, by comparing similar cases or by comparing cases that are visually similar, but with different diagnoses.

Comparison of Two Image Segmentation Algorithms

Image segmentation has been, and still is, a relevant research area in computer vision and hundreds of segmentation algorithms have been proposed in the last 30 years. An image retrieval system has now an array of available algorithm choices, however, few objective numerical evaluations of these segmentation algorithms exist. Typically, the effectiveness of a new algorithm is demonstrated only by the presentation of a few segmented images and an expert decides the effectiveness of a technique based on application requirements. This paper presents a comparison between two image segmentation algorithms: the color set back-projection algorithm that can be found in many related studies and an original segmentation method using a hexagonal structure defined on the set of image pixels. For this purpose error measures are used. The error measures quantify the consistency between these two segmentations algorithms.

A Topic Map for “Subject-Centric” Learning

The paper presents an original algorithm for automated building of a topic map starting from a relational database. This aspect is favorable in e-learning domain because many e-learning systems use a relational database. The process is illustrated on a database used in TESYS e-learning system. The reason for topicmap building and topic map graphical viewing is to offer students along with traditional learning modality named “course-centric” a new modality named “subject-centric”. This newmodality based on topicmaps allows learner to navigate through topicmap depending on its interest subject or to follow the associations between subjects. Students had the possibility to use these two learning modalities and they emphasized positive aspects of using topicmaps, for example the possibility to understand better the relationships of knowledge.

A graphical interface for evaluating three graph-based image segmentation algorithms

Image segmentation has an essential role in image Analysis, pattern recognition and low-level vision. Since multiple segmentation algorithms exists in literature, numerical evaluations are needed to quantify the consistency quantification because are allowing a principled comparison between segmentation results on different images, with differing numbers of regions, and generated by different algorithms with different parameters. This paper presents a graphical interface for evaluating three graph-based image segmentation algorithms: the color set back-projection algorithm, an efficient graph based image segmentation algorithm known also as the local variation algorithm and a new and original segmentation algorithm using a hexagonal structure defined on the set of image pixels.

Custom ontologies for an automated image annotation system

Automated annotation of digital images is a challenging task being used for indexing, retrieving, and understanding of large collections of image data. Several machine learning approaches have been proposed to model the existing associations between words and images. Each approach is trying to assign to a test image some meaningful words taking into account a set of feature vectors extracted from that image. In general for the annotation process of medical or natural images the words are retrieved from a controlled vocabulary or from an ontology. This paper presents an original approach for creating two ontologies and an original design of an image annotation system. The ontologies are created using the information provided by two distinct sources: MeSH - a vocabulary used for subject indexing and searching of journal articles in the life sciences and SAIAPR TC-12 Dataset - a set of annotated images having a vocabulary with a hierarchical structure. The annotation system is using an efficient annotation model called Cross Media Relevance Model each image being segmented using a segmentation algorithm based on a hexagonal structure.

Automated image annotation system based on an open source object database

Automated annotation of digital images is a challenging task being used for indexing, retrieving, and understanding of large collections of image data. Several machine-learning approached have been proposed to model the existing associations between words and images. Each approach is trying to assign to a test image some meaningful words taking into account a set of feature vectors extracted from that image. This paper presents an original image annotation system based on an open source object database called db4o. An object oriented model offers suport for storing complex objects as sets, lists, trees or other advanced data structures. The information needed for the annotation process is retrieved from the SAIAPR TC-12 Dataset - a set of annotated images having a vocabulary with a hierarchical structure. The annotation system is using an efficient annotation model called Cross Media Relevance Model.

Multimedia Elements for Medical e-Learning Improvement

The paper presents a software tool based on topic maps dedicated to medical e-learning. The application allows the graphical visualization of the MeSH thesaurus (medical terms from Diseases and Drugs categories) with a topic map. With this graphical modality, the learner can view medical term description and its associative relationships with other medical descriptors. The paper presents also how to use the topic map for semantic querying of a multimedia database with medical images that are accompanied by diagnosis and treatment as crucial information. For retrieving the interest information this access path can be combined with another modern solution: the content-based visual query on the multimedia medical database using color and texture features automatically extracted. The paper presents the original algorithm for building and populating the topic map starting from MeSH thesaurus, mapping an xml file that can be downloaded free, to an xtm file that contains the topic map.

Medical Imagistic Database Query for Educational Purpose

The paper presents original query modalities on a multimedia database that stores medical images and the associated information, for educational goal. So, a modern and efficient system for professional accomplishment is offered to the medical superior education (including residents, young specialists, family doctors and medical assistants). Specialists can update the medical image database with images acquired from different patients in the diagnosis and treatment process. A series of alphanumerical information: diagnosis, treatment and patient evolution can be added for each image. The database can be browsed, simply text-based queried or content-based queried using colour and texture characteristics automatically extracted from medical images at their loading in the database. An original element is the presence of a topic map based on a part of MeSH thesaurus, the part that includes the medical diagnosis names. The student can navigate through topic map depending on its interest subject, having in this way big advantages. He does not have to be familiar with the logic of the database, he will learn about the semantic context, in which a collection and its single items are embedded and he may find useful items he would not have expected to find in the beginning. Also, semantic queries against the multimedia database can be automatically launched with the help of the topic map. All these access paths can be combined for retrieving the interest information. Using content-based visual query with other access methods on a teaching image database allows students to see images and associated information from database in a simple and direct manner. This method stimulates learning, by comparing similar cases along with their particularities, or by comparing cases that are visually similar, but with different diagnoses.