Made available in DSpace on 2014-06-12T16:00:19Z (GMT). No. of bitstreams: 2 arquivo6372_1.pdf: 2413746 bytes, checksum: f2dce473f73c4dff94e0ee8f8ad4fec9 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2007 / Conselho Nacional de Desenvolvimento Científico e Tecnológico / Osteoporosis is a disease characterized by a reduction in bone mineral density and disruption of bone microarchitecture. It is the most common of the metabolic bone diseases, affecting more than 10,000,000 Brazilians. Despite being of paediatric origin, its symptoms are rarely manifested in childhood. However, the illness becomes increasingly problematic as the individual grows older, causing serious geriatric symptoms. As life expectancy increases, geriatric illnesses become a predominant factor in health care. An accurate diagnosis is therefore extremely important. Currently, the methods of diagnosis of osteoporosis are relatively efficient, albeit substantially inaccurate. For example, people with similar characteristics may have different diagnoses. By using a database of bone mineral density samples, this work presents a mathematical modeling able to produce a more accurate and less ambiguous diagnosis of osteoporosis. Several techniques of mathematical modeling were grouped around this objective. With the model in hand, this work raises new questions for the researcher concerned with the relationship between medicine and computer science. This work may be extended to solve similar problems in the medical domain, as well as in oceanography, meteorology, and others