Related Experiment Video
Updated: Aug 10, 2025
09:16
Multi-Modal Signals for Analyzing Pain Responses to Thermal and Electrical Stimuli
Published on: April 5, 2019
10.8K
Automatic assessment of pain based on deep learning methods: A systematic review.
Stefanos Gkikas1, Manolis Tsiknakis1
1Department of Electrical and Computer Engineering, Hellenic Mediterranean University, Estavromenos, Heraklion, 71410, Greece; Computational BioMedicine Laboratory, Institute of Computer Science, Foundation for Research & Technology-Hellas, Vassilika Vouton, Heraklion, 70013, Greece.
Computer Methods and Programs in Biomedicine
|February 10, 2023
Summary
Multimodal deep learning approaches significantly improve automatic pain assessment, especially when incorporating temporal data. This review highlights their importance for clinical pain management and suggests future research directions.
Area of Science:
- Artificial Intelligence
- Biomedical Engineering
- Pain Management
Background:
- Automatic pain assessment is crucial for effective pain management and preventing patient functional decline.
- Deep learning algorithms are increasingly used to analyze the complex nature of pain.
- This review synthesizes current research on deep learning for automatic pain assessment.
Approach:
- A systematic review of 110 publications was conducted.
- Studies were identified through Scopus, IEEE Xplore, and ACM Digital Library.
- Data were analyzed based on unimodal/multimodal approaches and temporal dimension usage.
Key Points:
- Multimodal approaches are vital for accurate automatic pain estimation, particularly in clinical settings.
- Integrating temporal data significantly enhances pain assessment performance.
- The review identifies effective deep learning architectures and learning methods.
Conclusions:
- Future research should focus on robust evaluation protocols and interpretable AI models.
- Limitations in current pain databases hinder optimal deep learning model development and validation.
- Enhanced deep learning models can serve as valuable decision-support tools in real-world clinical scenarios.
