Creating an image database containing mammography, ultrasound and elastography for the segmentation of suspicious findings

• Grant number: 15/17302-5
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Effective date (Start): November 26, 2015
• Effective date (End): September 25, 2016
• Field of knowledge: Engineering - Biomedical Engineering - Bioengineering
• Principal Investigator: Homero Schiabel
• Grantee: Karem Daiane Marcomini
• Supervisor: Robert Nishikawa
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Research place: University of Pittsburgh (Pitt), United States
• Associated to the scholarship: 12/24006-5 - Characterization of breast nodules in digital images of ultrasonography, elastography and mammography using intelligent techniques, BP.DR

Abstract

The screening techniques are very important for the early detection of breast cancer. The main difficulties are: subjectivity of the specialist to determine the lesion characteristics (boundaries and texture, for example) and when they the lesions are not seen in certain image modality, such as radiographically dense breasts in mammography or when there are the occurrence of overlapping tissues. This requires making other imaging exams, such as ultrasound and elastography. In addition, professionals in the technology area have proposed computational tools to improve the image quality, contrast enhancement, segmentation and classification of lesions and breast artifacts acting as a second opinion to the specialist. In this context, the proposal of this project is to create a large database containing mammography, B-mode ultrasound and elastography also register them in the system that is being developed in the PhD project. Another premise refers to aid of radiologists in the location of the lesions, draw its borders and description of the physiological aspects in order to assist in computational classification, being these the great contributions among research groups involved. The aim is also to apply pre-processing, segmentation and post-processing techniques which is already being studied, to evaluate the accuracy of this procedure and, if necessary, to study new automatic segmentation techniques.