WEB ENVIRONMENT FOR LEARNING AND TRAINING OF GRADUANTS, RESIDENTS OR PHYSICIANS IN THE INTERPRETATION OF BREAST IMAGES

• Grant number: 17/14016-7
• Support Opportunities: Regular Research Grants
• Duration: April 01, 2019 - September 30, 2021
• Field of knowledge: Interdisciplinary Subjects
• Principal Investigator: Silvia Cristina Martini Rodrigues
• Grantee: Silvia Cristina Martini Rodrigues
• Host Institution: Pró-Reitoria Acadêmica. Universidade de Mogi das Cruzes (UMC). Campus da Sede Mogi das Cruzes. Mogi das Cruzes , SP, Brazil
• Associated researchers: Marcia Aparecida Silva Bissaco

Abstract

Breast cancer is the most common type of cancer among women in the world and in Brazil, after non-melanoma skin, accounting for about 25% of new cases each year, but because this still happens if there is a screening plan Of this disease throughout the national territory? One of the hypotheses is the error in the interpretation of a mammographic image, since it is known that of each 9 women submitted to biopsy only one will have as diagnosis a true positive. When detected in time this disease has a cure rate of approximately 100%, but for this detection to be more precise in addition to more advanced equipment and with periodic quality control of both the equipment and the procedure is also necessary the investment in the analysis of the mammographic images. Based on the values of false negatives/false positives and the advance of new technologies is the proposal of this project. In this project will be developed an application to be attached to the Web environment under development by the group LAPSIM (Laboratory of Processing of Medical Images), called by BreastShare. However, focusing on e-learning for students in the health area, mainly for medical courses, Resident students or even to specialists who are taking a course of improvement in the fellow molds. This application will include learning contents, tests and activities, the most important being the comparison of the marking of structures in the mammographic image of the student with that of the teacher in real time. The system will consist of a breast imaging database with breast density types classified and marked by the specialist (doctor / tutor) or online database. The planning will be based on ISO / IEC 12207 or 15504. Through the browser (desktop or notebook), the user will visualize the breast images and perform the tasks requested by the teacher / tutor (mouse or touch screen), such as markings In the images, being that in this first stage of the project will be the marking of the region of mammary density. After marking, this will be compared to the gold standard already existing in the image base, and then the system will inform if the area demarcation has been set or not, this tool can be used as a type of teaching evaluation. This online markup is the novelty of this project, reducing the time the teacher / doctor / tutor would spend in correcting the evaluation. The validation will be done with the applicability of the environment in the medical course with qualitative and quantitative evaluations by area specialists and students. The project also meets the real needs of the radiology teacher in providing more training to their students. With this, it is believed that a better-trained physician will have a decrease in false positive and false negative diagnoses. This will have a direct impact on government spending, since breast cancer is the leading cause of retirement among neoplasms in Brazil (7% of total social security expenditures). What has been observed is a great increase in the access to the diagnosis due to the great extension to the access to mammography, but little has been done in search of an assertive diagnosis and this is done in the best capacity of the responsible person in the interpretation of the images. (AU)