Authors:
Carmen Maidantchik
1
;
José Manoel de Seixas
1
;
Afrânio Kritski
1
;
Fernanda C. de Q Mello
1
;
Rony T. V. Braga
1
;
Pedro H. S. Antunes
1
and
João Baptista de Oliveira e Souza Filho
2
Affiliations:
1
Federal University of Rio de Janeiro, Brazil
;
2
Federal University of Rio de Janeiro; Celso Suckow Technological Education Center, Brazil
Keyword(s):
Decision support systems, neural networks, web technology, SNPT diagnosis.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Artificial Intelligence and Decision Support Systems
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Computational Intelligence
;
Data Engineering
;
Enterprise Information Systems
;
Health Engineering and Technology Applications
;
Health Information Systems
;
Human-Computer Interaction
;
Information Systems Analysis and Specification
;
Knowledge Management
;
Methodologies and Methods
;
Neural Network Software and Applications
;
Neural Networks
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Ontologies and the Semantic Web
;
Pattern Recognition
;
Physiological Computing Systems
;
Sensor Networks
;
Signal Processing
;
Society, e-Business and e-Government
;
Soft Computing
;
Strategic Decision Support Systems
;
Theory and Methods
;
Web Information Systems and Technologies
Abstract:
The World Health Organization estimates that one third of the world population is infected by mycobacterium tuberculosis. Tuberculosis (TB) affects mainly poor health places in developing countries. Therefore, it became mandatory to develop more efficient, fast, and inexpensive analysis methods. This paper presents a decision support system that uses neural networks to sustain TB diagnosis. The output is the probability that a patient has or not the illness and an assigned risk group. The NeuralTB system encapsulates the knowledge needed for efficient anamnesis interview integrated to demographic and threat factors typically known for tuberculosis diagnosis. It was developed with the Web technology and data were described with a markup language to enable an efficient communication and information exchange among experts. Data collected during the whole process can be used to identify possible new factors or symptoms, since the infection transmission may evolve. This information can al
so support tuberculosis control governmental entities to define effective actions to protect the health and safety of the population.
(More)