Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 16, 2025Frontiers in Medical TechnologyOpen Access

Automated identification of early to mid-stage Parkinson’s disease using deep convolutional neural networks on static facial images

View Full Paper
Ask AI
Bookmark
Share

Authors

NYNi YangJLJing LiuLWLin Wang

Discussion

Loading...

Member takes

Overview

This analysis applies deep convolutional neural networks for detecting early-stage Parkinson's disease, indicating a non-invasive monitoring approach.

Key Points

  • ResNet18 demonstrated the highest performance with an F1 score of 99.67%, indicating exceptional accuracy.
  • Data augmentation expanded the sample size from 2,000 to 6,000 facial images, supporting robust model training.
  • Five different CNN architectures were fine-tuned, including noteworthy results from MobileNetV3 and EfficientNetV2.
  • Grad-CAM heatmaps identified critical facial regions around the eyes, lips, and nose as indicators for Parkinson's disease.

Cite This Study

Yang et al. (2025) studied this question.

synapsesocial.com/papers/68d4538731b076d99fa58c4ehttps://doi.org/10.3389/fmedt.2025.1655199
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. 1Innovative Deep Learning Approach for Parkinson's Disease Prediction: Leveraging Convolutional Neural Networks for Early Detection2024
  2. 2Enhancing Parkinson’s Disease Prediction Using Deep Learning-Based Convolutional Neural Networks2024
  3. 3Multimodal Retinal Imaging Classification for Parkinson's Disease Using a Convolutional Neural Network2024 · 8 citations
  4. 4An approach for Parkinson's disease detection based on Artificial Inteligence techniques2024 · 1 citations
  5. 5Parkinson’s Disease Prediction Using Convolutional Neural Networks and Hand-Drawn Image Analysis2024 · 5 citations