Related Experiment Video
Updated: Jun 29, 2026
07:15
Robust Ligature-Induced Model of Murine Periodontitis for the Evaluation of Oral Neutrophils
Published on: January 21, 2020
12.1K
DePerio: Innovative Deep Learning-Based Framework for Periodontal Disease Diagnosis and Severity Evaluation Using
IEEE Journal of Biomedical and Health Informatics
|December 23, 2025
Summary
Salivary oral polymorphonuclear neutrophils (oPMNs) can detect periodontal disease (PD) early. An AI pipeline, DePerio, accurately identifies and quantifies oPMNs in saliva, enabling timely diagnosis and monitoring.
Area of Science:
- Biomarkers and Diagnostics
- Oral Health Research
- Artificial Intelligence in Medicine
Background:
- Periodontal disease (PD) diagnosis often misses early stages.
- Salivary oral polymorphonuclear neutrophils (oPMNs) are promising early biomarkers.
- Existing deep learning models need adaptation for oPMN detection.
Purpose of the Study:
- To develop an AI pipeline (DePerio) for oPMN detection and quantification.
- To enable early diagnosis and monitoring of periodontal disease.
- To create a scalable platform for clinical dental practice.
Main Methods:
- Developed a novel oPMN isolation protocol.
- Integrated deep neural network (DNN) architectures for oPMN detection.
- Validated the pipeline against standard quantification methods for oPMNs and oral inflammatory load (OIL).
Main Results:
- DePerio achieved a detection error rate below 5%.
- The AI pipeline successfully classified five distinct levels of oral inflammatory load (OIL).
- Clinical study involved 111 human saliva samples from healthy to severe periodontitis cases.
Conclusions:
- DePerio offers a robust, low-complexity platform for early PD detection.
- The AI pipeline facilitates longitudinal monitoring of periodontal disease.
- This technology provides a practical solution for clinical dental settings.
