The Feasibility of Early Alzheimer’s Disease Diagnosis Using a Neural Network Hybrid Platform

Yu, Xinke; Srivastava, Siddharth; Huang, Shan; Hayden, Eric Y.; Teplow, David B.; Xie, Ya‐Hong

doi:10.3390/bios12090753

Cited by 9 publications

(10 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, due to insufficient literature included, the AUC value could not be calculated, which is also worth further exploration. In the application of different diagnostic models, we found that there were differences in the sensitivity and specificity of different models, this study found that there are diagnostic models based on CNN ( 15 , 30 ), LDA ( 16 , 17 , 19 , 20 ), SVM ( 18 ), MLP ( 21 ) and ANN ( 22 ), and the AUC value of LDA used in more studies was 0.934, which was not much different from that of pooled AUC, which seemed to suggest that LDA was suitable for the diagnosis and recognition of Raman spectroscopy in AD. Due to the limited number of other types of diagnostic models, it is not possible to calculate the AUC of other types of diagnostic models.…”

Section: Discussionmentioning

confidence: 73%

Accuracy of Raman spectroscopy in the diagnosis of Alzheimer's disease

Pan

et al. 2023

Front. Psychiatry

View full text Add to dashboard Cite

ObjectiveTo systematically evaluate the accuracy of Raman spectroscopy in the diagnosis of Alzheimer's disease.MethodsDatabases including Web of Science, PubMed, The Cochrane Library, EMbase, CBM, CNKI, Wan Fang Data, and VIP were electronically searched for studies on Raman spectroscopy in diagnosis of Alzheimer's disease from inception to November 2022. Two reviewers independently screened the literature, extracted data, and assessed the risk of bias in the included studies. Then, meta-analysis was performed using Meta-Disc1.4 and Stata 16.0 software.ResultsA total of eight studies were finally included. The pooled sensitivity of Raman spectroscopy was 0.86 [95% CI (0.80–0.91)], specificity was 0.87 [95% CI (0.79–0.92)], positive likelihood ratio was 5.50 [95% CI (3.55–8.51)], negative likelihood ratio was 0.17 [95% CI (0.09–0.34)], diagnosis odds ratio and area under the curve of SROC were 42.44 [95% CI (19.80–90.97)] and 0.931, respectively. Sensitivity analysis was carried out after each study was excluded one by one, and the results showed that pooled sensitivity and specificity had no significant change, indicating that the stability of the meta-analysis results was great.ConclusionsOur findings indicated that Raman spectroscopy had high accuracy in the diagnosis of AD, though it still did not rule out the possibility of misdiagnosis and missed diagnosis. Limited by the quantity and quality of the included studies, the above conclusions need to be verified by more high-quality studies.

show abstract

Section: Discussionmentioning

confidence: 73%

Accuracy of Raman spectroscopy in the diagnosis of Alzheimer's disease

Pan

et al. 2023

Front. Psychiatry

View full text Add to dashboard Cite

show abstract

“…The authors employed the ADNI dataset for the experimental study and achieved 99.79% accuracy. X. Yu et al [20] introduced a new method for identifying AD. They applied CNN with surface-enhanced Raman Spectroscopy (SERS) fingerprints of human Cerebrospinal Fluid (CSF).…”

Section: A Analysis Of Admentioning

confidence: 99%

“…Structural MRI [6], [20], [26], [34], [35], [40] GARD Investigation 4: How does the strategic optimization of hyperparameters influence the convergence and generalization capabilities of DL models intended for ADidentifying biomarkers? PS: The optimization of hyperparameters is vital in the construction of DL models for the discovery of AD biomarkers.…”

Section: Figure 5 Ad Detection Using Various Technique Modelsmentioning

confidence: 99%

Adversarial Network-Based Classification for Alzheimer’s Disease Using Multimodal Brain Images: A Critical Analysis

Gupta,

Kumar,

Abraham

2024

IEEE Access

View full text Add to dashboard Cite

Alzheimer's disease (AD) is a progressive neurodegenerative disorder that represents a significant and growing public health challenge. This work concisely summarizes AD, encompassing its pathophysiology, risk factors, clinical manifestations, diagnosis, treatment, and ongoing research. The main goal of managing AD is to reduce symptoms while improving the lives of those impacted. This letter has conducted a systematic review to analyze the prediction of AD using the Preferred Reporting Item for Systematic Review and Meta-Analysis (PRISMA) guidelines. The major scientific databases such as Scopus, Web of Science (WoS), and IEEE Xplorer are explored, where 2018-2023 publications are considered. The article selection process is based on keywords like "Alzheimer's disease," "Brain Images," "Deep Learning (DL)," etc. After rigorous analysis, 946 articles were extracted, and 42 were identified for final consideration. Further, several investigations based on the previous work are discussed along with its Proposed Solutions (PS). Finally, a case study on AD detection using the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset and AD Detection Network (ADD-NET) implementation is presented.

show abstract

“…15 Xie group investigated the self-association and LODs of Aβ peptides in detail 16,17 and further demonstrated a convolutional neural network-based AD diagnosis approach using the SERS fingerprints of human cerebrospinal fluid (CSF). 18 Compared with conventional disease diagnosis methods such as ELISA, chromatography, and mass spectrometry, SERS technology has the advantages of high sensitivity and convenience of detection and analysis, which makes it highly promising in the early diagnosis of some intractable diseases such as AD. 15 High-performance substrate paves the way for the application of SERS technology.…”

Section: Introductionmentioning

confidence: 99%

“…Park et al detected the AD markers Tau protein and Aβ42 peptide using three-dimensional (3D) gold nanowire arrays prepared by nanoimprinting as solid SERS substrates and obtained the limit of detections (LODs) of 10 –11 moles per liter (M) and 10 –9 M of the two biomarkers, respectively, corresponding to a SERS enhancement factor (EF) of 5.5 × 10 5 . Xie group investigated the self-association and LODs of Aβ peptides in detail , and further demonstrated a convolutional neural network-based AD diagnosis approach using the SERS fingerprints of human cerebrospinal fluid (CSF) . Compared with conventional disease diagnosis methods such as ELISA, chromatography, and mass spectrometry, SERS technology has the advantages of high sensitivity and convenience of detection and analysis, which makes it highly promising in the early diagnosis of some intractable diseases such as AD …”

Section: Introductionmentioning

confidence: 99%

Label-Free Analysis of Protein Biomarkers Using Pattern-Optimized Graphene-Nanopyramid SERS for the Rapid Diagnosis of Alzheimer’s Disease

Wu,

Duan,

Jiang

et al. 2024

ACS Appl. Nano Mater.

View full text Add to dashboard Cite

The quantitative and highly sensitive detection of biomarkers such as Tau proteins and Aβ polypeptides is considered one of the most effective methods for the early diagnosis of Alzheimer's disease (AD). Surface-enhanced Raman spectroscopy (SERS) detection is a promising method that faces, however, challenges like insufficient sensitivity due to the non-optimized nanostructures for specialized analyte sizes and insufficient control of the location of SERS hot spots. Thus, the SERS detection of AD biomarkers is restricted. We reported here an indepth study of the analytical Raman enhancement factor (EF) of the wafer-scale graphene-Au nanopyramid hybrid SERS substrates using a combination of both theoretical calculation and experimental measurements. Experimental results show that larger nanopyramids and smaller gap spacing lead to a larger SERS EF, with an optimized analytical EF up to 1.1 × 10 10 . The hybrid SERS substrate exhibited detection limits of 10 −15 M for Tau and phospho-Tau (P-Tau) proteins and 10 −14 M for Aβ polypeptides, respectively. Principal component analysis correctly categorized the SERS spectra of different biomarkers at ultralow concentrations (10 −13 M) using the optimized substrate. Amide III bands at 1200−1300 cm −1 reflect different structural conformations of proteins or polypeptides. Tau and P-Tau proteins are inherently disordered with a few α-helix residuals. The structure of Aβ42 polypeptides transitioned from the α-helix to the β-sheet as the concentration increased. These results demonstrate that the hybrid SERS method could be a simple and effective way for the label-free detection of protein biomarkers to enable the rapid early diagnosis of AD and other diseases.

show abstract

The Feasibility of Early Alzheimer’s Disease Diagnosis Using a Neural Network Hybrid Platform

Cited by 9 publications

References 44 publications

Accuracy of Raman spectroscopy in the diagnosis of Alzheimer's disease

Accuracy of Raman spectroscopy in the diagnosis of Alzheimer's disease

Adversarial Network-Based Classification for Alzheimer’s Disease Using Multimodal Brain Images: A Critical Analysis

Label-Free Analysis of Protein Biomarkers Using Pattern-Optimized Graphene-Nanopyramid SERS for the Rapid Diagnosis of Alzheimer’s Disease

Contact Info

Product

Resources

About