Classification techniques in cardiac diagnosis based on motion quantification of nuclear medicine images

• Grant number: 05/04614-7
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: July 01, 2006
• End date: June 30, 2009
• Field of knowledge: Engineering - Biomedical Engineering
• Principal Investigator: Roberto Hirata Junior
• Grantee: Carlos da Silva dos Santos
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

This project fits into a interdisciplinar research program on statistical and computational learning methods for classification of medical images and signals. Our main goal is investigating innovative and robust computational learning methods for aiding medical decision based on images. As secondary goals, we intend: (1) to create an integrated computational environment for analysis and classification of 3D images using open source tools; (2) to investigate representations of classifiers which are more fit for the interpretation of the clinical professional. Statistical learning tools for diagnosis assistance have as benefit the reduction of time for the task, increased repeatability and the possibility of reusing the results. We intend to combine the knowledge of the clinical specialist with learning tools to deal with the excessive data present in 3D image sequences. Other aspect to be adressed is the use of innovative techniques for compacting the information which is used for classification. The classifyers to be implemented will be based on data extracted from tomographic images, using methods for measuring the cardiac movement.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

SANTOS, CARLOS S.; HIRATA, NINA S. T.; HIRATA, ROBERTO. An Information Theory framework for two-stage binary image operator design. PATTERN RECOGNITION LETTERS, v. 31, n. 4, p. 10-pg., 2010-03-01. (05/04614-7)

SANTOS, CARLOS S.; HIRATA, NINA S. T.; HIRATA, ROBERTO. An Information Theory framework for two-stage binary image operator design. PATTERN RECOGNITION LETTERS, v. 31, n. 4, SI, p. 297-306, MAR 1 2010. (05/04614-7)

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

SANTOS, Carlos da Silva dos. Binary feature extraction based on interaction analysis. 2010. Doctoral Thesis - Universidade de São Paulo (USP). Instituto de Matemática e Estatística (IME/SBI) São Paulo.