Saulo Domingos de Souza Pedro
Research Subjects
Research Interests
Saulo Domingos de Souza Pedro has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Augusto Braga Fernandes Antunes
Alliar - NEPIA
2 shared publications - Bruno Machado Pacheco
Federal University of Santa Catarina (UFSC)
2 shared publications - Danilo Silva
UFSC
2 shared publications For The Alzheimer’s Disease Neuroimaging Initiative
1 shared publicationVictor H. R. de Oliveira
Federal University of Santa Catarina (UFSC)
1 shared publication- Victor Hugo Rocha de Oliveira
Federal University of Santa Catarina (UFSC) Florianopolis SC Brazil
1 shared publication
Latest Publications
- 12 October 2023
- Published by Springer Nature
- p. 181-194
Does Pre-training on Brain-Related Tasks Results in Better Deep-Learning-Based Brain Age Biomarkers?
- 12 October 2023
- Published by Springer Nature
- p. 181-194
Does Pre-training on Brain-Related Tasks Results in Better Deep-Learning-Based Brain Age Biomarkers?
- Open Access
Does pre-training on brain-related tasks results in better deep-learning-based brain age biomarkers?
Brain age prediction using neuroimaging data has shown great potential as an indicator of overall brain health and successful aging, as well as a disease biomarker. Deep learning models have been established as reliable and efficient brain age estimators, being trained to predict the chronological age of healthy subjects. In this paper, we investigate the impact of a pre-training step on deep learning models for brain age prediction. More precisely, instead of the common approach of pre-training on natural imaging classification, we propose pre-training the models on brain-related tasks, which led to state-of-the-art results in our experiments on ADNI data. Furthermore, we validate the resulting brain age biomarker on images of patients with mild cognitive impairment and Alzheimer's disease. Interestingly, our results indicate that better-performing deep learning models in terms of brain age prediction on healthy patients do not result in more reliable biomarkers.
Open AccessDoes pre-training on brain-related tasks results in better deep-learning-based brain age biomarkers?
Brain age prediction using neuroimaging data has shown great potential as an indicator of overall brain health and successful aging, as well as a disease biomarker. Deep learning models have been established as reliable and efficient brain age estimators, being trained to predict the chronological age of healthy subjects. In this paper, we investigate the impact of a pre-training step on deep learning models for brain age prediction. More precisely, instead of the common approach of pre-training on natural imaging classification, we propose pre-training the models on brain-related tasks, which led to state-of-the-art results in our experiments on ADNI data. Furthermore, we validate the resulting brain age biomarker on images of patients with mild cognitive impairment and Alzheimer's disease. Interestingly, our results indicate that better-performing deep learning models in terms of brain age prediction on healthy patients do not result in more reliable biomarkers.