2022
DOI: 10.21037/atm-22-2961
|View full text |Cite
|
Sign up to set email alerts
|

Diagnostic accuracy study of automated stratification of Alzheimer’s disease and mild cognitive impairment via deep learning based on MRI

Abstract: Background: Alzheimer's disease (AD) is a widespread neurodegenerative disease that mostly affects the elderly population. Given its prevalence, a precise and efficient stratification system based on AD symptomology that uses functional magnetic resonance imaging (MRI) has great potential in the clinical diagnosis and prognosis estimation of AD patients. It was evident that deep learning methods have performed extremely well in the field of automated stratification of AD based on MRI because of their high pred… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
7
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 10 publications
(7 citation statements)
references
References 45 publications
0
7
0
Order By: Relevance
“…To further evaluate the performance of our DMA-HPCNet model, we conducted a comprehensive comparison with various advanced studies, as shown in Table VII, which includes network feature extraction forms, dataset sizes, and experimental results. Based on the network feature extraction forms, we summarised recent research on AD classification tasks, including two region-level methods [12], [14], two patch-level methods [10], [20], and two slice-level methods [19], [24]. From Table VII, several conclusions were drawn as follows.…”
Section: Results Analysismentioning
confidence: 99%
See 2 more Smart Citations
“…To further evaluate the performance of our DMA-HPCNet model, we conducted a comprehensive comparison with various advanced studies, as shown in Table VII, which includes network feature extraction forms, dataset sizes, and experimental results. Based on the network feature extraction forms, we summarised recent research on AD classification tasks, including two region-level methods [12], [14], two patch-level methods [10], [20], and two slice-level methods [19], [24]. From Table VII, several conclusions were drawn as follows.…”
Section: Results Analysismentioning
confidence: 99%
“…Most of these methods involve extracting 3D information to model, which is difficult to achieve in a lightweight manner. It is worth noting that our DMA-HPCNet model selects cross sections from 3D MRI images after registration to construct a lightweight CNN model, which is superior to other slice-level methods [19], [24] in AD/MCI and MCI/CN classification tasks. Finally, based on these results, we conclude that our proposed DMA-HPCNet model, which uses hybrid pyramid convolution block and dual multi-level attention mechanism, proves its validity in AD diagnosis with excellent performance.…”
Section: Results Analysismentioning
confidence: 99%
See 1 more Smart Citation
“…In the current studies on AD model building, few studies have validated the screened genes through in vitro experiments, mostly through another dataset or GC patients ( Chen W. et al, 2021 ; Liu et al, 2021 ; Chen et al, 2022 ). Chen W. et al (2021) successfully constructed an AD prediction model using the ADNI database and combining clinical and imaging histological features, however, it was not validated by in vitro experiments.…”
Section: Discussionmentioning
confidence: 99%
“…ML techniques were utilized in a wide range of software, including image identifi cation, clinical issue, and categorization if it would be diffi cult or impractical to perform the necessary procedures using different algorithms (7). ML is thought to be a feature of AI that imitates the human mind analyzes data and develops correlations to facilitate the decision (8). AI systems can learn unsupervised through unprocessed or unstructured information to Deep Learning (DL) is a subtype of computer vision.…”
Section: Introductionmentioning
confidence: 99%