Related Experiment Video
Updated: Jan 17, 2026
09:16
Multi-Modal Signals for Analyzing Pain Responses to Thermal and Electrical Stimuli
Published on: April 5, 2019
11.4K
Feasibility and Effectiveness of a Low-Code AI Platform for Developing a Neonatal Multimodal Pain Classification
Nannan Yang1, Xiaosong Jiang2, Xue Jin2
1Department of Nursing, Nanjing Drum Tower Hospital Clinical College of Nanjing University of Chinese Medicine, Nanjing, Jiangsu, People's Republic of China.
Journal of Multidisciplinary Healthcare
|September 19, 2025
Summary
Low-code AI platforms enable effective neonatal pain classification model development. This approach bridges the research-to-practice gap, empowering clinicians to create accessible AI tools for improved infant pain management.
Area of Science:
- Neonatal care
- Artificial intelligence in medicine
- Machine learning for healthcare
Background:
- Artificial intelligence (AI) shows promise in neonatal pain recognition, but clinical application is limited by complex algorithms.
- Low-code AI development platforms offer a solution by simplifying AI model creation for practical use.
Purpose of the Study:
- To assess the feasibility of building and validating a neonatal multimodal pain classification model using a low-code AI platform (EasyDL).
- To create an accessible, cost-effective method for clinicians to develop AI tools without advanced programming skills.
Main Methods:
- A neonatal multimodal pain dataset (426 segments) was used to train a video classification model on the EasyDL platform via AutoML.
- The model underwent internal testing and external validation against N-PASS scores, with prospective data.
Main Results:
- The AI model achieved 89.6% accuracy and 85.8% F1 score in internal validation.
- External validation showed 87.7% accuracy and AUC > 0.95 for all pain levels.
- The model was successfully deployed via API to an Android device for clinical use.
Conclusions:
- Developing neonatal multimodal pain classification models with low-code AI is feasible and effective.
- This approach facilitates AI democratization, enabling clinicians to create AI solutions for neonatal pain management.
