期刊
COMPUTERS IN BIOLOGY AND MEDICINE
卷 136, 期 -, 页码 -出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.compbiomed.2021.104725
关键词
Bayesian deep learning; Diabetic retinopathy; Lesion segmentation; Microaneurysm; Hard exudate; Soft exudate; Haemorrhage
类别
资金
- LUT Doctoral School
- CSC - IT Center for Science, Finland
This study extends recent research on deep neural networks with a Bayesian framework, considering parameters as random variables and using stochastic variational dropout for uncertainty quantification. The proposed method improves lesion segmentation accuracy and analyzes model uncertainty, which is crucial for early detection of abnormalities.
Early diagnosis of retinopathy is essential for preventing retinal complications and visual impairment due to diabetes. For the detection of retinopathy lesions from retinal images, several automatic approaches based on deep neural networks have been developed in the recent years. Most of the proposed methods produce point estimates of pixels belonging to the lesion areas and give no or little information on the uncertainty of method predictions. However, the latter can be essential in the examination of the medical condition of the patient when the goal is early detection of abnormalities. This work extends the recent research with a Bayesian framework by considering the parameters of a convolutional neural network as random variables and utilizing stochastic variational dropout based approximation for uncertainty quantification. The framework includes an extended validation procedure and it allows analyzing lesion segmentation distributions, model calibration and prediction uncertainties. Also the challenges related to the deep probabilistic model and uncertainty quantification are presented. The proposed method achieves area under precision-recall curve of 0.84 for hard exudates, 0.641 for soft exudates, 0.593 for haemorrhages, and 0.484 for microaneurysms on IDRiD dataset.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据