Related Experiment Video
Updated: May 9, 2025
05:56
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
Published on: April 14, 2023
2.3K
Artificial Intelligence in Oral Diagnosis: Detecting Coated Tongue with Convolutional Neural Networks
Sümeyye Coşgun Baybars1, Merve Hacer Talu1, Çağla Danacı2
1Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Firat University, Elazığ 23119, Turkey.
Diagnostics (Basel, Switzerland)
|May 1, 2025
Summary
A new AI model using Support Vector Machine and VGG19 deep learning accurately detects coated tongue lesions. This tool aids in diagnosing this common oral condition, potentially indicating underlying health issues.
Area of Science:
- Oral Medicine
- Artificial Intelligence
- Medical Imaging
Background:
- Coated tongue is a common oral condition often overlooked due to its asymptomatic nature.
- It can indicate poor oral hygiene and signal underlying systemic diseases.
- Early detection is crucial for timely intervention and management.
Purpose of the Study:
- To develop a robust diagnostic model for coated tongue detection.
- To utilize convolutional neural networks (CNNs) and machine learning (ML) classifiers for improved lesion identification.
- To enhance clinical decision support for oral health assessment.
Main Methods:
- Analysis of 200 tongue images (100 coated, 100 healthy) captured with a DSLR camera.
- Feature extraction using CNN architectures (VGG16, VGG19, ResNet, MobileNet, NasNet).
- Classification using Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Multi-Layer Perceptron (MLP) models.
Main Results:
- The hybrid SVM + VGG19 model achieved the highest performance.
- Achieved sensitivity of 82.6%, specificity of 88.23%, accuracy of 85%, and F1 score of 86.36%.
- Demonstrated superior diagnostic capability compared to other tested configurations.
Conclusions:
- The SVM + VGG19 model is accurate and reliable for diagnosing coated tongue.
- This AI-driven approach shows potential as a clinical decision support tool.
- Further research with larger datasets can improve model generalizability across diverse populations.
