Abstract
Alzheimer’s disease (AD) is the most common progressive neurodegenerative disease. 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) is widely used to predict AD using a deep learning model. However, the effects of noise and blurring on 18F-FDG PET images were not considered. The performance of a classification model trained using raw, deblurred (by the fast total variation deblurring method), or denoised (by the median modified Wiener filter) 18F-FDG PET images without or with cropping around the limbic system area using a 3D deep convolutional neural network was investigated. The classification model trained using denoised whole-brain 18F-FDG PET images achieved classification performance (0.75/0.65/0.79/0.39 for sensitivity/specificity/F1-score/Matthews correlation coefficient (MCC), respectively) higher than that with raw and deblurred 18F-FDG PET images. The classification model trained using cropped raw 18F-FDG PET images achieved higher performance (0.78/0.63/0.81/0.40 for sensitivity/specificity/F1-score/MCC) than the whole-brain 18F-FDG PET images (0.72/0.32/0.71/0.10 for sensitivity/specificity/F1-score/MCC, respectively). The 18F-FDG PET image deblurring and cropping (0.89/0.67/0.88/0.57 for sensitivity/specificity/F1-score/MCC) procedures were the most helpful for improving performance. For this model, the right middle frontal, middle temporal, insula, and hippocampus areas were the most predictive of AD using the class activation map. Our findings demonstrate that 18F-FDG PET image preprocessing and cropping improves the explainability and potential clinical applicability of deep learning models.
Highlights
Alzheimer’s disease (AD) is a progressive neurodegenerative disease characterized by cognitive decline and memory loss [1], and it is the most common cause of dementia, which causes disability and dependency in older people worldwide [2]
This study aimed to investigate: (1) whether applying the denoising and deblurring method to 18F-FDG positron emission tomography (PET) images improves the performance of AD classification, and (2) whether 18F-FDG PET image cropping improves the performance of AD classification using a modified deep learning model from 3D-ResNet, which has been recently described as a powerful prediction model for 3D medical images [16]
Our findings provide evidence that preprocessing the input features facilitates clinical decision-making and interpretation of the classification model based on 18F-FDG PET images
Summary
Alzheimer’s disease (AD) is a progressive neurodegenerative disease characterized by cognitive decline and memory loss [1], and it is the most common cause of dementia, which causes disability and dependency in older people worldwide [2]. Cognitive decline may be associated with metabolic and neurotransmitter activities in the brain [3]. These changes in AD may start several years before the onset of clinical symptoms [4,5]. Since the earliest symptoms of AD, such as short-term memory loss, are confused with symptoms resulting from aging, stress, or other brain disorders, it remains challenging to recognize AD before the manifestation of severe cognitive impairment with typical neuroimaging signs [7]. A recognized approach to detect AD in the early stages is urgently needed
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call