Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 23, 2026PLoS ONEOpen Access

Feature integration of 18FFDG PET brain imaging using deep learning for sensitive cognitive decline detection

View Full Paper
Ask AI
Bookmark
Share

Authors

YLYeji LeeSKS. H. KimSKSangil Kim

Discussion

Loading...

Member takes

Overview

Randomized trial finds improved cognitive decline detection using integrated PET features, indicating enhanced diagnostic accuracy.

Key Points

  • The aim is to improve the detection of cognitive decline, including early Alzheimer's disease, using advanced imaging techniques.
  • Voxel-level features extracted using CNN and PCANet from 252 participants' PET imaging.
  • Region-level features obtained from standardized uptake value ratios processed by a deep neural network.
  • Classification models were trained and validated with 5-fold cross-validation, comparing performance against MMSE scores.
  • The integrated DNN-CNN model achieved the highest accuracy of 0.87 ± 0.05, showing a 6.33% improvement over DNN-only.
  • Recall increased from 0.77 to 0.88, and F1-Score improved from 0.82 to 0.88 after integration.
  • Predicted probabilities of cognitive decline significantly correlated with MMSE scores, surpassing MMSE-based classification accuracy.

Cite This Study

Lee et al. (2026) studied this question.

synapsesocial.com/papers/6a61af8bfaa9903c5116a4d6https://doi.org/10.1371/journal.pone.0341995
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Union is strength: the combination of radiomics features and 3D-deep learning in a sole model increases diagnostic accuracy in demented patients: a whole brain 18FDG PET-CT analysis2024 · 3 citations
  2. 218F-FDG-PET-based deep learning for predicting cognitive decline in non-demented elderly across the Alzheimer’s disease clinical spectrum2024 · 3 citations
  3. 3Introducing an ensemble method for the early detection of Alzheimer's disease through the analysis of PET scan images2024 · 1 citations
  4. 4Classification of Alzheimer's diseases from PET Images Using a Convolutional Neural Network2024 · 4 citations
  5. 5Cognition-Weighted Multimodal MRI and FDG-PET Features for Classification of Mild Cognitive Impairment, Alzheimer’s Disease, and Frontotemporal Dementia2026