Deep Learning for Detecting Dental Plaque and Gingivitis From Oral Photographs: A Systematic Review
Bibliographic record
Abstract
OBJECTIVES: This systematic review aimed to evaluate the performance of deep learning (DL) models in detecting dental plaque and gingivitis from red, green, and blue (RGB) intraoral photographs. METHODS: A comprehensive literature search was conducted across Medline, Scopus, Embase, and Web of Science databases up to January 31, 2025. The methodological characteristics and performance metrics of studies developing and validating DL models for classification, detection, or segmentation tasks were analysed. The risk of bias was assessed using the quality assessment of diagnostic accuracy studies 2 (QUADAS-2) tool, and the certainty of the evidence was evaluated with the grading of recommendations assessment, development, and evaluation (GRADE) framework. RESULTS: From 3307 identified records, 23 studies met the inclusion criteria. Of these, 10 focused on dental plaque, 11 on gingivitis, and two addressed both outcomes. The risk of bias was low in all QUADAS-2 domains for 11 studies, with low applicability concerns in nine. For dental plaque, DL models showed robust performance in the segmentation task, with intersection over union (IoU) values ranging from 0.64 to 0.86 (median 0.74). Three studies indicated that DL models outperformed dentists in identifying dental plaque when disclosing agents were not used. For gingivitis, the models demonstrated potential but underperformed compared to dental plaque, with IoU values ranging from 0.43 to 0.72 (median 0.63). The certainty of the evidence was moderate for dental plaque and low for gingivitis. CONCLUSIONS: DL models demonstrate promising potential for detecting dental plaque and gingivitis from intraoral photographs, with superior performance in plaque detection. Leveraging accessible imaging devices such as smartphones, these models can enhance teledentistry and may facilitate early screening for periodontal disease. However, the lack of external testing, multicenter studies, and reporting consistency highlights the need for further research to ensure real-world applicability.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.012 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.007 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".