Banca de QUALIFICAÇÃO: NICKERSON FONSECA FERREIRA

Uma banca de QUALIFICAÇÃO de DOUTORADO foi cadastrada pelo programa.
STUDENT : NICKERSON FONSECA FERREIRA
DATE: 17/12/2024
TIME: 14:00
LOCAL: Google Meet, https://meet.google.com/zks-zzpn-ujk
TITLE:

Development of an Intelligent Tool for Breast Cancer Diagnosis Support Using Federated Learning

KEY WORDS:

Breast Cancer; Machine Learning; Federated Learning.

PAGES: 29
BIG AREA: Ciências Biológicas
AREA: Biologia Geral
SUMMARY:

Breast cancer is the most commonly diagnosed form of the disease and one of the leading causes of death among women. However, if diagnosed early, the mortality Programa de Pos-Graduacao em Bioinformatica - PPg-Bioinfo http://www.posgraduacao.ufrn.br/bioinfo Contato: (84) 3342-2216 (ramal 123) / (84) 99480-6818 (WhatsApp) bioinfo@imd.ufrn.br Endereco: Instituto do Cerebro (por tras do IMD). Campus Universitario Central da UFRN - Av. Cap. Mor Gouveia, S/N - Lagoa Nova, Natal - RN, 59078-900 rate can be significantly reduced. With the increasing technological advances, especially in the area of artificial intelligence, the use of tools to assist healthcare professionals in the detection and diagnosis of the disease is becoming increasingly common in institutions around the world. The present work has the objective of developing an intelligent tool where its central purpose is to reduce the cancer diagnosis time through the detection of breast lesions and the creation of an intelligent queue that prioritizes mammograms with a higher risk of detected malignancy. For this, a machine learning model will be developed responsible for classifying the images obtained through mammograms, based on BI-RADS labels. In order to increase efforts in the fight against cancer, the concept of federated learning will be applied. This technique allows a network of healthcare institutions to train their models locally and then share the weights for a central (global) model, ensuring data privacy and increasing the generalization of the global model. The intelligent queue will manage pending mammogram reports, taking into account the classification made by the machine learning model and the time this exam is awaiting analysis. With this, there will be a reduction in the time it takes to issue reports for exams with a higher probability of cancer, which is crucial for treatment and mortality reduction

COMMITTEE MEMBERS:
Presidente - 3083298 - RENAN CIPRIANO MOIOLI
Interno - 3884005 - PATRICK CESAR ALVES TERREMATTE
Externo à Instituição - ANDRÉ PERES