Abstract
Alzheimer's Disease (AD) is the most common neurodegenerative disease, with 10% prevalence in the elder population. Conventional Machine Learning (ML) was proven effective in supporting the diagnosis of AD, while very few studies investigated the performance of deep learning and transfer learning in this complex task. In this paper, we evaluated the potential of ensemble transfer-learning techniques, pretrained on generic images and then transferred to structural brain MRI, for the early diagnosis and prognosis of AD, with respect to a fusion of conventional-ML approaches based on Support Vector Machine directly applied to structural brain MRI. Specifically, more than 600 subjects were obtained from the ADNI repository, including AD, Mild Cognitive Impaired converting to AD (MCIc), Mild Cognitive Impaired not converting to AD (MCInc), and cognitively-normal (CN) subjects. We used T1-weighted cerebral-MRI studies to train: (1) an ensemble of five transfer-learning architectures pretrained on generic images; (2) a 3D Convolutional Neutral Network (CNN) trained from scratch on MRI volumes; and (3) a fusion of two conventional-ML classifiers derived from different feature extraction/selection techniques coupled to SVM. The AD-vs-CN, MCIc-vs-CN, MCIc-vs-MCInc comparisons were investigated. The ensemble transfer-learning approach was able to effectively discriminate AD from CN with 90.2% AUC, MCIc from CN with 83.2% AUC, and MCIc from MCInc with 70.6% AUC, showing comparable or slightly lower results with the fusion of conventional-ML systems (AD from CN with 93.1% AUC, MCIc from CN with 89.6% AUC, and MCIc from MCInc with AUC in the range of 69.1–73.3%). The deep-learning network trained from scratch obtained lower performance than either the fusion of conventional-ML systems and the ensemble transfer-learning, due to the limited sample of images used for training. These results open new prospective on the use of transfer learning combined with neuroimages for the automatic early diagnosis and prognosis of AD, even if pretrained on generic images.
Highlights
With an estimate of 5.7 million people affected in 2018 in the only United States and a prevalence of 10% in the elder population [> 65 years old, [1]], Alzheimer’s Disease (AD) is the most common neurodegenerative disease, accounting for 50–75% of all cases of dementia [2].To date, AD can be definitely diagnosed only after death, with post-mortem examinations aimed at measuring the presence of amyloid plaques and neurofibrillary tangles
The 3D Convolutional Neutral Network (CNN) reached convergence within 80 epochs with an AUC of 84.1% in classifying AD vs. CN, 72.3 for Mild Cognitive Impaired converting to AD (MCIc) vs. CN and 61.1 for MCIc vs. Mild Cognitive Impaired not converting to AD (MCInc), lower than the one obtained by both conventional Machine Learning (ML) and 2D transfer learning, due to the limited sample of images used for training
When considering the MCIcvs-MCInc discrimination, the performance of fused conventional MLs improve from 69.1% up to 73.3% using the inner cerebral structures instead of the entire Magnetic Resonance Imaging (MRI) volume, while those obtained from the inner structures by the ensemble transfer-learning remain stable at 70.6%
Summary
With an estimate of 5.7 million people affected in 2018 in the only United States and a prevalence of 10% in the elder population [> 65 years old, [1]], Alzheimer’s Disease (AD) is the most common neurodegenerative disease, accounting for 50–75% of all cases of dementia [2].To date, AD can be definitely diagnosed only after death, with post-mortem examinations aimed at measuring the presence of amyloid plaques and neurofibrillary tangles. A probable or possible clinical diagnosis of AD is often mainly based on patient’s self-reported experiences and the assessment of behavioral, functional, and cognitive status through neuropsychological tests and questionnaires This approach results to be insufficient for the diagnosis of AD, especially in the early pre-dementia stage of the disease known as Mild Cognitive Impairment (MCI), whose rate of progression to Alzheimer’s dementia is only 33% [4]. Due to these weaknesses and according to many scientific evidences arising in the last years, the revised diagnostic criteria for AD published in 2011 included neuroimaging studies as techniques able to detect signs of the disease even before dementia is apparent [5, 6]. These methods can provide measurements of AD-specific proteins’ deposit and reduced metabolism/atrophic regions, respectively, related to the presence and the progression of AD [7, 8]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
