Ontology Based SWRL Rules for Diagnostic of Tumoral Bone Pathologies

Abstract

Bone cancer is one of the deadliest cancers in the world. It grows in the skeletal system and destroys tissue. It can spread to adjacent organs, such as the lungs, and occurs when a tumor or abnormal tissue mass forms in a bone. A tumor may be malignant, which means it’s growing aggressively and spreading to other parts of the body. This article deals with the diagnostic process of bone tumors. In order to analyze a big volume of medical data, ontologies are the most efficient technique to improve medical image analysis used to detect different tumors and other bone lesions. Therefore. The main objective is to show the contribution of semantic reasoning coupled with the ontological model to detect and diagnose bone cancer disease. The essential characteristics of our approach are the diagnosis of bone tumors through SWRL inference rules. The major advantage of this work is essentially to integrate the reasoning into our Ontobone ontology modeled in a previous work in order to assist in the decision-makinort phase in terms of diagnosis, risk estimation and the proposal of appropriate treatments whatever for the treatment of tumoral bone pathologies or to prevent their risks. The evaluation of our work was based on a set of clinical cases from the medical folder of patients from the radiological service of the CHU Hedi Chaker of Sfax and which have system have correctly diagnosed 27 out of the 40 patients (ratio of correctness is approximately around 90%). These results suggest that the proposed approach could be useful for staging and processing using classification systems. Additionally, we have developed a prototype OntoBone system that demonstrates the effectiveness of our proposed approach.