Multiscale deep neural network based analysis of FDG-PET images for the early diagnosis of Alzheimer’s disease

• Published in: Medical image analysis Vol. 46; pp. 26 - 34
• Main Authors: Lu, Donghuan, Popuri, Karteek, Ding, Gavin Weiguang, Balachandar, Rakesh, Beg, Mirza Faisal
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.05.2018; Elsevier BV
• Subjects: Aged; Alzheimer Disease - diagnostic imaging; Alzheimer's disease; Artificial neural networks; Brain; Classifiers; Cognitive ability; Dementia disorders; Early Diagnosis; Female; Fluorodeoxyglucose F18; Glucose; Humans; Impairment; Male; Medical diagnosis; Medical imaging; Memory; Metabolism; Metabolism FDG-PET; Multiscale analysis; Multiscale deep neural network learning; Neural networks; Neural Networks (Computer); Neurodegenerative diseases; Neuroimaging; Neurological diseases; Positron emission; Positron emission tomography; Positron-Emission Tomography - methods; Radiopharmaceuticals; Tomography; Early diagnosis; Multiscale deep neural network learning; Alzheimer’s disease; Metabolism FDG-PET
• ISSN: 1361-8415; 1361-8423; 1361-8423
• DOI: 10.1016/j.media.2018.02.002

•Proposed a novel multiscale deep neural network to learn the patterns of metabolism changes due to AD pathology as discriminative from the patterns of metabolism in normal controls (NC).•Showed that by transferring samples from NC and AD individuals, the deep architecture can obtain better discriminative ability in the early diagnosis task.•Demonstrated that ensemble multiple classifiers with different validation settings can make the proposed method more stable and robust, and improve its classification performance.•We present a comprehensive validation of our method analyzing metabolism measures taken from 1051 subjects that were processed with stringent quality control requirements including expert manual editing of all segmentations for ensuring accuracy. To-date, our study is perhaps the first study to utilize such a large number of FDG-PET images, and hence, these results indicate good potential for generalizability. [Display omitted] Alzheimer’s disease (AD) is one of the most common neurodegenerative diseases with a commonly seen prodromal mild cognitive impairment (MCI) phase where memory loss is the main complaint progressively worsening with behavior issues and poor self-care. However, not all individuals clinically diagnosed with MCI progress to AD. A fraction of subjects with MCI either progress to non-AD dementia or remain stable at the MCI stage without progressing to dementia. Although a curative treatment of AD is currently unavailable, it is extremely important to correctly identify the individuals in the MCI phase that will go on to develop AD so that they may benefit from a curative treatment when one becomes available in the near future. At the same time, it would be highly desirable to also correctly identify those in the MCI phase that do not have AD pathology so they may be spared from unnecessary pharmocologic interventions that, at best, may provide them no benefit, and at worse, could further harm them with adverse side-effects. Additionally, it may be easier and simpler to identify the cause of the cognitive impairment in these non-AD cases, and hence proper identification of prodromal AD will be of benefit to these individuals as well. Fluorodeoxy glucose positron emission tomography (FDG-PET) captures the metabolic activity of the brain, and this imaging modality has been reported to identify changes related to AD prior to the onset of structural changes. Prior work on designing classifier using FDG-PET imaging has been promising. Since deep-learning has recently emerged as a powerful tool to mine features and use them for accurate labeling of the group membership of given images, we propose a novel deep-learning framework using FDG-PET metabolism imaging to identify subjects at the MCI stage with presymptomatic AD and discriminate them from other subjects with MCI (non-AD / non-progressive). Our multiscale deep neural network obtained 82.51% accuracy of classification just using measures from a single modality (FDG-PET metabolism data) outperforming other comparable FDG-PET classifiers published in the recent literature.