Abstract Alzheimer Research Articles

Lipopolysaccharide distinctively alters human microglia transcriptomes to resemble microglia from Alzheimer's disease mouse models.

ABSTRACTAlzheimer's disease (AD) is the most common form of dementia, and risk-influencing genetics implicates microglia and neuroimmunity in the pathogenesis of AD. Induced pluripotent stem cell (iPSC)-derived microglia (iPSC-microglia) are increasingly used as a model of AD, but the relevance of historical immune stimuli to model AD is unclear. We performed a detailed cross-comparison over time on the effects of combinatory stimulation of iPSC-microglia, and in particular their relevance to AD. We used single-cell RNA sequencing to measure the transcriptional response of iPSC-microglia after 24 h and 48 h of stimulation with prostaglandin E2 (PGE2) or lipopolysaccharide (LPS)+interferon gamma (IFN-γ), either alone or in combination with ATPγS. We observed a shared core transcriptional response of iPSC-microglia to ATPγS and to LPS+IFN-γ, suggestive of a convergent mechanism of action. Across all conditions, we observed a significant overlap, although directional inconsistency to genes that change their expression levels in human microglia from AD patients. Using a data-led approach, we identify a common axis of transcriptomic change across AD genetic mouse models of microglia and show that only LPS provokes a transcriptional response along this axis in mouse microglia and LPS+IFN-γ in human iPSC-microglia. This article has an associated First Person interview with the first author of the paper.

Label-free hyperspectral imaging and deep-learning prediction of retinal amyloid β-protein and phosphorylated tau.

Alzheimer’s disease (AD) is a major risk for the aging population. The pathological hallmarks of AD—an abnormal deposition of amyloid β-protein (Aβ) and phosphorylated tau (pTau)—have been demonstrated in the retinas of AD patients, including in prodromal patients with mild cognitive impairment (MCI). Aβ pathology, especially the accumulation of the amyloidogenic 42-residue long alloform (Aβ42), is considered an early and specific sign of AD, and together with tauopathy, confirms AD diagnosis. To visualize retinal Aβ and pTau, state-of-the-art methods use fluorescence. However, administering contrast agents complicates the imaging procedure. To address this problem from fundamentals, ex-vivo studies were performed to develop a label-free hyperspectral imaging method to detect the spectral signatures of Aβ42 and pS396-Tau, and predicted their abundance in retinal cross-sections. For the first time, we reported the spectral signature of pTau and demonstrated an accurate prediction of Aβ and pTau distribution powered by deep learning. We expect our finding will lay the groundwork for label-free detection of AD.

Yeni Tiyazolil-Hidrazin Türevlerinin Sentezi ve Asetilkolinesteraz (AChE) ve Bütirilkolinesteraz (BuChE) Aktivite Çalışmaları

Alzheimer hastalığı (AH), zamanla beyin hücrelerinin ölümüne bağlı olarak bilişsel işlevlerde azalma, hafıza kaybı ve bunama ile ilişkilendirilmiştir. Alzheimer hastalığının tedavisinde sınırlayıcı uygulamaların mevcut oluşu, hastalığın sağlık harcamalarında kanser ve kalp hastalıklardan sonra dünyada üçüncü sırada oluşu araştırmacıları AH üzerinde erken evrede tanıma ve yeni tedavi yöntemlerine yönlendirmektedir. Günümüzde kolinerjik anomalikler ile AH arasında doğrudan bir ilişki olduğu düşünülmektedir. Çalışmalar, asetilkolinesteraz (AChE) ve bütirilkolinesteraz (BuChE) inhibisyonunun asetilkolin (ACh) seviyesinde meydana gelen artışların Alzheimer hastalığının başlangıç evrelerindeki bilişsel yetmezliği iyileştirebileceğini kanıtlamıştır. Dolayısıyla ACh düzeylerini arttırmak için uygulanacak en iyi metot ise, ACh’yi yıkan AChE veya BuChE enzimlerinin baskılanmasıdır. Dolayısıyla yapılan bu çalışmada, sübstitüe edilmiş tiyazolilhidrazin türevleri tasarlanmış, sentezlenmiş ve AH’ ye karşı asetilkolinesteraz (AChE) ve bütirilkolinesteraz (BuChE) kolinesteraz enzimlerinin inhibisyon potansiyelleri araştırılmıştır. Hedef bileşiklerin yapıları 1H NMR/13C NMR analiz yöntemleri ile aydınlatılmıştır. Hedef bileşiklerin AChE ve BuChE enzimleri üzerindeki inhibisyon etkileri Ellman yöntemiile değerlendirilmiş ve hedef bileşiklerin enzim inhibisyon çalışmaları sonucunda 3d bileşiğinin orta düzeyde bütirilkolinesteraz enzimini inhibe ettiği tespit edilmiştir.

Discriminant Subspace Low-Rank Representation Algorithm for Electroencephalography-Based Alzheimer's Disease Recognition.

Alzheimer’s disease (AD) is a chronic progressive neurodegenerative disease that often occurs in the elderly. Electroencephalography (EEG) signals have a strong correlation with neuropsychological test results and brain structural changes. It has become an effective aid in the early diagnosis of AD by exploiting abnormal brain activity. Because the original EEG has the characteristics of weak amplitude, strong background noise and randomness, the research on intelligent AD recognition based on machine learning is still in the exploratory stage. This paper proposes the discriminant subspace low-rank representation (DSLRR) algorithm for EEG-based AD and mild cognitive impairment (MCI) recognition. The subspace learning and low-rank representation are flexibly integrated into a feature representation model. On the one hand, based on the low-rank representation, the graph discriminant embedding is introduced to constrain the representation coefficients, so that the robust representation coefficients can preserve the local manifold structure of the EEG data. On the other hand, the least squares regression, principle component analysis, and global graph embedding are introduced into the subspace learning, to make the model more discriminative. The objective function of DSLRR is solved by the inexact augmented Lagrange multiplier method. The experimental results show that the DSLRR algorithm has good classification performance, which is helpful for in-depth research on AD and MCI recognition.

Modelling the Effects of Medium-Chain Triglycerides on Cerebral Ketone Body Metabolism

Alzheimer’s Disease (AD) is a neurodegenerative disorder that causes drastic structural brain atrophy and affects multiple brain functions. Cerebral glucose hypometabolism, associated with senile plaque density formation, is a pre-symptomatic feature of AD and significantly contributes to AD’s future development and progression. As cerebral glucose metabolism gradually slows down due to advanced aging, a healthy adult brain experiences an 8% decrease in cerebral glucose metabolic rate (CGMR) compared to a decline of 20%–40% CGMR in AD patients. To bridge the increasing brain energy gap caused by glucose hypometabolism, ketone bodies (KBs) are used as a supplementary source of energy as cerebral KB metabolism remains unaffected in AD patients. Ketogenic interventions such as Medium-Chain Triglyceride (MCT)-induced treatment can help augment the brain’s energy source availability and might delay further cognitive decline. With this, we constructed a mathematical model on cerebral glucose and KB metabolism to illustrate the drastic effects of glucose hypometabolism on healthy aging individuals, Mild Cognitive Impairment (MCI) subjects, and AD patients. Through the generated simulations, we have shown that KB concentration levels rise during prolonged starvation, and in consideration of glucose hypometabolism, MCT-induced intervention increases the concentration levels of acetyl-CoA (AC) in MCI/AD patients. Furthermore, MCT-induced supplement helps increase the AC concentration levels in healthy adults under normal conditions.

Photo-Oxygenation as a New Therapeutic Strategy for Neurodegenerative Proteinopathies by Enhancing the Clearance of Amyloid Proteins.

Alzheimer disease (AD) is associated with the aggregation of two amyloid proteins: tau and amyloid-β (Aβ). The results of immunotherapies have shown that enhancing the clearance and suppressing the aggregation of these two proteins are effective therapeutic strategies for AD. We have developed photocatalysts that attach oxygen atoms to Aβ and tau aggregates via light irradiation. Photo-oxygenation of these amyloid aggregates reduced their neurotoxicity by suppressing their aggregation both in vitro and in vivo. Furthermore, photo-oxygenation enhanced the clearance of Aβ in the brain and microglial cells. Here, we describe the effects of photo-oxygenation on tau and Aβ aggregation, and the potential of photo-oxygenation as a therapeutic strategy for AD, acting via microglial clearance.

Diterpenoid Caesalmin C Delays Aβ-Induced Paralysis Symptoms via the DAF-16 Pathway in Caenorhabditis elegans.

Alzheimer’s disease (AD) is the most prevalent neurodegenerative disease in the world. However, there is no effective drug to cure it. Caesalmin C is a cassane-type diterpenoid abundant in Caesalpinia bonduc (Linn.) Roxb. In this study, we investigated the effect of caesalmin C on Aβ-induced toxicity and possible mechanisms in the transgenic Caenorhabditis elegans AD model. Our results showed that caesalmin C significantly alleviated the Aβ-induced paralysis phenotype in transgenic CL4176 strain C. elegans. Caesalmin C dramatically reduced the content of Aβ monomers, oligomers, and deposited spots in AD C. elegans. In addition, mRNA levels of sod-3, gst-4, and rpt-3 were up-regulated, and mRNA levels of ace-1 were down-regulated in nematodes treated with caesalmin C. The results of the RNAi assay showed that the inhibitory effect of caesalmin C on the nematode paralysis phenotype required the DAF-16 signaling pathway, but not SKN-1 and HSF-1. Further evidence suggested that caesalmin C may also have the effect of inhibiting acetylcholinesterase (AchE) and upregulating proteasome activity. These findings suggest that caesalmin C delays the progression of AD in C. elegans via the DAF-16 signaling pathway and that it could be developed into a promising medication to treat AD.

Why Women Are Predisposed to Alzheimer’s Disease

Alzheimer’s disease (AD) exhibits progressive dementia, profound disability, and impaired quality of life. Globally, there are more than 50 million AD patients, over two-thirds of which are women. The cause behind this gender difference had remained unclear for decades until FSH (follicle stimulating hormone), a hormone that dramatically increases in women after menopause, was recently found to trigger this phenomenon.

Tau Acts in Concert With Kinase/Phosphatase Underlying Synaptic Dysfunction.

Alzheimer's disease (AD) is characterized by two pathological features: neurofibrillary tangles (NFTs), formed by microtubule-associated protein tau, and abnormal accumulation of amyloid-β (Aβ). Multiple evidence placed synaptic tau as the vital fact of AD pathology, especially at the very early stage of AD. In the present review, we discuss tau phosphorylation, which is critical for the dendritic localization of tau and synaptic plasticity. We review the related kinases and phosphatases implicated in the synaptic function of tau. We also review the synergistic effects of these kinases and phosphatases on tau-associated synaptic deficits. We aim to open a new perspective on the treatment of AD.

Predicting the Effects of Repetitive Transcranial Magnetic Stimulation on Cognitive Functions in Patients With Alzheimer's Disease by Automated EEG Analysis.

Alzheimer's disease (AD) is a progressive, neurodegenerative brain disorder that generally affects the elderly. Today, after the limited benefit of the pharmacological treatment strategies, numerous noninvasive brain stimulation techniques have been developed. Transcranial magnetic stimulation (TMS), based on electromagnetic stimulation, is one of the most widely used methods. The main problem in the use of TMS is the existence of large individual variability in the results. This causes a waste of money, time, and more importantly, a burden for delicate patients. Hence, it is a necessity to form an efficient and personalized TMS application protocol. In this paper, we performed a machine-learning analysis to see whether it is possible to predict the responses of patients with AD to TMS by analyzing their electroencephalography (EEG) signals. For that purpose, we analyzed both the EEG signals collected before and after the TMS application (EEG1 and EEG2, respectively). Through correlating EEG1 and repetitive transcranial magnetic stimulation (rTMS) outcomes, we tried to see whether it is possible to predict patients' responses before the treatment application. On the other hand, by EEG2 analysis, we investigated TMS impacts on EEG, more importantly if this impact is correlated with patients' response to the treatment. We used the support vector machine (SVM) classifier due to its multiple advantages for the current task with feature selection processes by stepwise linear discriminant analysis (SWLDA) and SVM. However, to justify our numerical analysis framework, we examined and compared the performances of different feature selection and classification techniques. Since we have a limited sample number, we used the leave-one-out method for the validation with the Monte Carlo technique to eliminate bias by a small sample size. In the conclusion, we observed that the correlation between rTMS outcomes and EEG2 is stronger than EEG1, since we observed, respectively, 93 and 79% of accuracies during our data analysis. Besides the informative features of EEG2 are focused on theta band, it indicates that TMS is characterizing the theta band signals in patients with AD in direct relation to patients' response to rTMS. This shows that it is more possible to determine patients' benefit from the TMS at the early stages of the treatment, which would increase the efficiency of rTMS applications in patients with Alzheimer's disease.

Transcriptional Profiling of Hippocampus Identifies Network Alterations in Alzheimer’s Disease

Alzheimer’s disease (AD) is a neurodegenerative disease characterized by rapid brain cell degeneration affecting different areas of the brain. Hippocampus is one of the earliest involved brain regions in the disease. Modern technologies based on high-throughput data have identified transcriptional profiling of several neurological diseases, including AD, for a better comprehension of genetic mechanisms of the disease. In this study, we investigated differentially expressed genes (DEGs) from six Gene Expression Omnibus (GEO) datasets of hippocampus of AD patients. The identified DEGs were submitted to Weighted correlation network analysis (WGCNA) and ClueGo to explore genes with a higher degree centrality and to comprehend their biological role. Subsequently, MCODE was used to identify subnetworks of interconnected DEGs. Our study found 40 down-regulated genes and 36 up-regulated genes as consensus DEGs. Analysis of the co-expression network revealed ACOT7, ATP8A2, CDC42, GAD1, GOT1, INA, NCALD, and WWTR1 to be genes with a higher degree centrality. ClueGO revealed the pathways that were mainly enriched, such as clathrin coat assembly, synaptic vesicle endocytosis, and DNA damage response signal transduction by p53 class mediator. In addition, we found a subnetwork of 12 interconnected genes (AMPH, CA10, CALY, NEFL, SNAP25, SNAP91, SNCB, STMN2, SV2B, SYN2, SYT1, and SYT13). Only CA10 and CALY are targets of known drugs while the others could be potential novel drug targets.

Autophagy Regulation Influences β-Amyloid Toxicity in Transgenic Caenorhabditis elegans.

Alzheimer’s disease (AD) is a progressive, neurodegenerative disease characterized by the accumulation of amyloid-beta (Aβ) proteins in the form of plaques that cause a proteostasis imbalance in the brain. Several studies have identified autophagy deficits in both AD patients and AD animal models. Here, we used transgenic Caenorhabditis elegans to study the relationship between autophagy flux and Aβ. We labeled autophagosomes with an advanced fluorescence reporter system, and used this to observe that human Aβ expression caused autophagosome accumulation in C. elegans muscle. The autophagy-related drugs chloroquine and 3-MA were employed to investigate the relationship between changes in autophagic flux and the toxicity of Aβ expression. We found that reducing autophagosome accumulation delayed Aβ-induced paralysis in the CL4176 strain of C. elegans, and alleviated Aβ-induced toxicity, thus having a neuroprotective effect. Finally, we used RNA-sequencing and proteomics to identify genes whose expression was affected by Aβ aggregation in C. elegans. We identified a series of enriched autophagy-related signal pathways, suggesting that autophagosome accumulation impairs Aβ protein homeostasis in nematodes. Thus, maintaining normal autophagy levels appears to be important in repairing the protein homeostasis imbalance caused by Aβ expression.

A pathological role of the Hsp40 protein Ydj1/DnaJA1 in models of Alzheimer's disease.

Alzheimer's disease (AD) is the most common form of dementia with millions of people affected worldwide. Pathophysiological manifestations of AD include the extracellular accumulation of amyloid beta (Abeta) pep-tides, products of the proteolytic cleavage of the amy-loid precursor protein APP. Increasing evidence sug-gests that Abeta peptides also accumulate intracellular-ly, triggering neurotoxic events such as mitochondrial dysfunction. However, the molecular factors driving formation and toxicity of intracellular Abeta are poorly understood. In our recent study [EMBO Mol Med 2022 - e13952], we used different eukaryotic model systems to identify such factors. Based on a genetic screen in yeast and subsequent molecular analyses, we found that both the yeast chaperone Ydj1 and its human ortholog DnaJA1 physically interact with Abeta, facili-tate the aggregation of Abeta peptides into small oli-gomers and promote their translocation to mitochon-dria. Deletion or downregulation of this chaperone pro-tected from Abeta-mediated toxicity in yeast and Dro-sophila AD models, respectively. Most importantly, the identified chaperone is found to be dysregulated in post-mortem human samples of AD patients. Here, we aim to outline our key findings, highlighting pathological functions of a heat shock protein (Hsp) family member, which are generally considered protective rather than toxic during neurodegeneration. Our results thus chal-lenge the concept of developing generalized chaperone activation-based therapies and call for carefully consid-ering also maladaptive functions of specific heat shock proteins.

Mechanism of NLRP3 inflammasome activation and its role in Alzheimer’s disease

Alzheimer’s disease (AD) is a common neurological disease in the elderly, and the major manifestations are cognitive dysfunction, neuronal loss, and neuropathic lesions in the brain. In the process of AD pathogenesis, the inflammatory response plays an indispensable role. The nucleotide-binding oligomerization domain (NOD)-like receptor family pyrin domain containing 3 (NLRP3) inflammasome containing NOD, leucine-rich repeat (LRR), and pyran domains is a multi-molecular complex that can detect dangerous signals related to neurological diseases. The assembly of NLRP3 inflammasome promotes the maturation of interleukin-1beta (IL-1β) and IL-18 mediated by caspase-1 in microglia, which leads to neuroinflammation and finally contributes to the occurrence and development of AD. This review aimed to clarify the structure and activating mechanism of NLRP3 inflammasome and its key role in the pathogenesis of AD, summarize the latest findings on the suppression of NLRP3 inflammasome activation for the treatment of AD, as well as indicate that targeting regulation of NLRP3 inflammasome assembly may be a potential strategy for the treatment of AD, providing a theoretical basis for the research of AD.

Artificial Intelligence Predictor for Alzheimer's Disease Trained on Blood Transcriptome: The Role of Oxidative Stress.

Alzheimer’s disease (AD) is an incurable neurodegenerative disease diagnosed by clinicians through healthcare records and neuroimaging techniques. These methods lack sensitivity and specificity, so new antemortem non-invasive strategies to diagnose AD are needed. Herein, we designed a machine learning predictor based on transcriptomic data obtained from the blood of AD patients and individuals without dementia (non-AD) through an 8 × 60 K microarray. The dataset was used to train different models with different hyperparameters. The support vector machines method allowed us to reach a Receiver Operating Characteristic score of 93% and an accuracy of 89%. High score levels were also achieved by the neural network and logistic regression methods. Furthermore, the Gene Ontology enrichment analysis of the features selected to train the model along with the genes differentially expressed between the non-AD and AD transcriptomic profiles shows the “mitochondrial translation” biological process to be the most interesting. In addition, inspection of the KEGG pathways suggests that the accumulation of β-amyloid triggers electron transport chain impairment, enhancement of reactive oxygen species and endoplasmic reticulum stress. Taken together, all these elements suggest that the oxidative stress induced by β-amyloid is a key feature trained by the model for the prediction of AD with high accuracy.

Impact of Chronic Conditions and Dementia in Rural West Texas: A Healthy Aging Study

Alzheimer’s disease (AD) is a devastating illness in elderly individuals, that currently has no known cure. Causal genetic factors only account for 1-2% of AD patients. However, other causal factors are still unknown for a majority of AD patients. Currently, multiple factors are implicated in late-onset AD, including unhealthy diet, physical inactivity, traumatic brain injury, chronic conditions, epigenetic factors, and environmental exposures. Although clinical symptoms of dementia are common to all races and ethnic groups, conditions that lead to dementia are different in terms of lifestyle, genetic profile, and socio-economic conditions. Increasing evidence also suggests that some elderly individuals age without cognitive impairments in their 60–90s as seen in rural West Texas, while some individuals progress with chronic conditions and cognitive impairments into their 60s. To understand these discriminations, we assessed current literature on demographic features of health in rural West Texas. This paper also outlines our initiated clinical study with a purpose of understanding the factors that allow some individuals to live without cognitive impairments at the age of 60–90 years, whereas others develop deficits in cognitive function around or above 60 years. Our ongoing study hopes to determine the factors that delay aging in some individuals by investigating various aspects including genetics, epigenetics, ethnicity, biology, culture, and lifestyle. This will be achieved by gathering information about participants’ ethnographic profiles, cognitive assessments, blood-profiles, brain scans, and blood-based genomic analyses in relation to lifestyle. The outcomes of our study will provide insights into healthy aging in rural West Texas.

Induced Pluripotent Stem Cell-Derived Neural Culture Model of Alzheimer’s Disease: A 3D Organoid Approach

Alzheimer’s disease (AD) is a progressive neurologic disorder that impacts a diverse population of older adults. As three-dimensional (3D) models are powerful tools for advancing AD studies, the authors have been developed AD cortical organoids to enable the observation of AD pathology at the cellular, tissue, and organ levels. For creating the model, APPSwe/Ind (APP) and PSEN1 (PS1) mutant genes were transfected into mouse induced pluripotent stem cells (iPSCs) following which the iPSC lines that expressed mutant APP and PS1 proteins were obtained. Then, using modified serum-free suspended embryoid body culture, AD cerebral organoids were made successfully at various ages. The AD model can show AD’s biochemical and pathological alterations, such as overexpressions of Aβ40 and Aβ42 and a decrease of GABAergic interneurons. The proposed model has the potential for implementation in many biomedical applications, including AD drug screening, stem cell transplant, and neuronal tissue engineering.

Trained Immunity in Neurodegenerative Diseases

Abstract Alzheimer’s disease (AD) and Parkinson’s disease (PD) are complex and progressive neurodegenerative disturbances leading to cognitive or motor dysfunctions and lacking effective therapy. A deregulation of innate immune response and inflammation appears to be deeply involved in both illnesses, though with still unclear mechanisms. Evidence in animals attributes a neurodegeneration-modulating role to mechanisms of innate memory, but human data on this topic are still scarce. In this study, we investigated whether -possibly due to concomitant factors including enhanced infectious burden, microbial dysbiosis and genetic risk factors linked to immunity-the innate immune cells of AD and PD patients show long-term functional reprogramming characterized by either potentiated or tolerant responses, which could ultimately contribute to disease development. Thus, following repeated in vitro stimulation with either heat-inactivated Candida albicans or LPS, we measured the cytokine response of CD14+ cells from AD and PD patients, both in comparison with those of age- and gender-matched healthy control subjects and in association with symptom severity. We observed different cytokine patterns associated with tolerant or potentiated responses among the three groups of subjects, pointing to different and disease-dependent trained immune status in patients. Our results would provide insights about how previous triggering of innate immunity can influence neuroinflammation and disease progression, offering new knowledge on trained immunity impact in neurodegenerative diseases.

Non-classical anti-inflammatory drugs (NCAIDs) ameliorates brain inflammation and improves memory in Alzheimer’s diseases mice model

Abstract Alzheimer’s disease (AD) is the most common cause of dementia due to inflammatory neurodegeneration in the brain. Neuroinflammation caused by damage associated molecular pattern (DAMP: Aβ, ATP) leads to neuronal apoptosis. From our study HY209, a NCAIDs and a TGR5 agonist, inhibits neuroinflammation and neurodegeneration. In vitro, HY209 suppressed NLRP3-ASC oligomerization and production of mature IL-1β/IL-18 in microglia cells by Aβ ± ATP. HY209 inhibited Ca++ influx incurred by ATP or BzATP (P2X7 agonists). Ca++ influx of microglia from TGR5 KO was like that of P2X7 KO compared to wild type mice in response to BzATP. Suggesting that TGR5 is essential in P2X7-mediated Ca++ signaling and HY209 might play a role as an inverse agonist for TGR5 colocalized with P2X7. HY209 activates adenylate cyclase and cAMP. cAMP inhibits priming phase of inflammasomal activation, suppresses pro-inflammatory cytokine (TNF-α) production and oligomerization of NLRP3-ASC complex. Indicating that HY209 suppresses both priming and activation phase of NLRP3 inflammasome. In the 5x Familial Alzheimer’s disease (5xFAD) mice model, HY209 (1 mpk, i.p. or 30 mpk, p.o.) improves memory function, decreases Aβ plaque and neuronal apoptosis in the brain. Structure activity relationships were studied to screen out more potent HY209 derivatives (next generation NCAIDs) with increased bioavailability and higher BBB permeability. Considering efficacies of NSAIDs, PPARγ agonists, specific inhibitors (for TNF-α, ROS, RAGE, HMG-CoA, p38 MAPK, P2X7 and NLRP3) targeting limited number of proinflammatory pathways, NCAIDs could ameliorate broad spectrum of neuroinflammation that is crucial in preventing neurodegeneration and improving cognition and memory in AD.

Fusing Multimodal and Anatomical Volumes of Interest Features Using Convolutional Auto-Encoder and Convolutional Neural Networks for Alzheimer's Disease Diagnosis.

Alzheimer’s disease (AD) is an age-related disease that affects a large proportion of the elderly. Currently, the neuroimaging techniques [e.g., magnetic resonance imaging (MRI) and positron emission tomography (PET)] are promising modalities for AD diagnosis. Since not all brain regions are affected by AD, a common technique is to study some region-of-interests (ROIs) that are believed to be closely related to AD. Conventional methods used ROIs, identified by the handcrafted features through Automated Anatomical Labeling (AAL) atlas rather than utilizing the original images which may induce missing informative features. In addition, they learned their framework based on the discriminative patches instead of full images for AD diagnosis in multistage learning scheme. In this paper, we integrate the original image features from MRI and PET with their ROIs features in one learning process. Furthermore, we use the ROIs features for forcing the network to focus on the regions that is highly related to AD and hence, the performance of the AD diagnosis can be improved. Specifically, we first obtain the ROIs features from the AAL, then we register every ROI with its corresponding region of the original image to get a synthetic image for each modality of every subject. Then, we employ the convolutional auto-encoder network for learning the synthetic image features and the convolutional neural network (CNN) for learning the original image features. Meanwhile, we concatenate the features from both networks after each convolution layer. Finally, the highly learned features from the MRI and PET are concatenated for brain disease classification. Experiments are carried out on the ADNI datasets including ADNI-1 and ADNI-2 to evaluate our method performance. Our method demonstrates a higher performance in brain disease classification than the recent studies.