RT Conference Proceedings
T1 Improving Uncertainty Estimations for Mammogram Classification using Semi-Supervised Learning
A1 Calderón-Ramírez, Saúl
A1 Murillo-Hernández, Diego
A1 Rojas-Salazar, Kevin
A1 Calvo-Valverde, Luis-Alexander
A1 Yang, Shengxiang
A1 Moemeni, Armaghan
A1 Elizondo Acuña, David Alberto
A1 López-Rubio, Ezequiel
A1 Molina-Cabello, Miguel Ángel
K1 Mamas - Cáncer - Diagnóstico
AB Computer aided diagnosis for mammogram images have seen positive results through the usage of deep learning architectures. However, limited sample sizes for the target datasets might prevent the usage of a deep learning model under real world scenarios. The usage of unlabeled data to improve the accuracy of the model can be an approach to tackle the lack of target data. Moreover, important model attributes for the medical domain as model uncertainty might be improved through the usage of unlabeled data. Therefore, in this work we explore the impact of using unlabeled data through the implementation of a recent approach known as MixMatch, for mammogram images.We evaluate the improvement on accuracy and uncertainty ofthe model using popular and simple approaches to estimateuncertainty. For this aim, we propose the usage of the uncertaintybalanced accuracy metric.
YR 2021
FD 2021-07
LK https://hdl.handle.net/10630/22699
UL https://hdl.handle.net/10630/22699
LA eng
NO Universidad de Málaga. Campus de Excelencia Internacional Andalucía Tech.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 20 ene 2026