RT Conference Proceedings
T1 Parkinsons Disease Detection by using Isosurfaces with Convolutional Neural Networks
A1 Ortiz-García, Andrés
K1 Parkinson, Enfermedad de
K1 Inteligencia artificial
K1 Congresos y conferencias
AB Computer aided diagnosis systems based on brain imaging are an important toolto assist in the diagnosis of Parkinson’s disease. The ultimate goal would be detec-tion by automatic recognizing of patterns that characterize the disease. In recenttimes Convolutional Neural Networks (CNN) have proved to be amazingly usefulfor that task. The drawback, however, is that 3D brain images contains a hugeamount of information that leads to complex CNN architectures. When thesearchitectures become too complex, classification performances often degrades be-cause the limitations of the training algorithm and overfitting. Thus, this paperproposes the use of isosurfaces as a way to reduce such amount of data whilekeeping the most relevant information. These isosurfaces are then used to im-plement a classification system which uses two of the most well-known CNNarchitectures to classify DaTScan images with an averageaccuracy of 95.1% and AUC=97%, obtaining comparable (slightly better) valuesto those obtained for most of the recently proposed systems. It can be concludedtherefore that the computation of isosurfaces reduces the complexity of the inputssignificantly, resulting in high classification accuracies with reduced computa-tional burden.
YR 2019
FD 2019
LK https://hdl.handle.net/10630/17864
UL https://hdl.handle.net/10630/17864
LA spa
NO Universidad de Málaga. Campus de Excelencia Internacional Andalucía Tech.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 21 ene 2026