A Novel Deep Learning Framework for Multi-Class Orthopantomogram-Derived Single-Tooth Disease Classification
Bibliographic record
Abstract
Dental radiographs are essential for diagnosing tooth-related diseases, yet their interpretation is often time-consuming and varies among clinicians.Although deep learning has advanced dental image analysis, most existing studies remain limited to binary classification or specific imaging modalities.This study aims to develop and evaluate a deep learning framework capable of multi-class single-tooth disease classification from radiographic images.A dataset of 4,439 single-tooth images was prepared from annotated dental radiographs, representing four clinically relevant categories: caries, deep caries, impacted teeth, and periapical lesions.The network integrates efficient convolutional and attention-based feature extraction with anti-aliased down-sampling and multi-scale feature aggregation to enhance representation and calibration reliability.Training employed a twophase augmentation strategy (heavy to light) and weighted cross-entropy loss under stratified five-fold cross-validation, with final predictions obtained through soft-voting ensemble averaging.The framework achieved an accuracy of 0.980 and an F1-score of 0.974, surpassing the performance reported in recent single-tooth classification studies, which typically achieve accuracies in the range of 0.92-0.95and F1-scores of approximately 0.83-0.94.These findings indicate that deep learning can provide accurate, consistent, and interpretable multi-class diagnosis at the tooth level, potentially reducing the diagnostic workload of dental professionals and allowing greater focus on complex clinical cases.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".