CLASEG: advanced multiclassification and segmentation for differential diagnosis of oral lesions using deep learning
Bibliographic record
Abstract
Oral cancer has a high mortality rate primarily due to delayed diagnoses, highlighting the need for early detection of oral lesions. This study presents a novel deep learning framework for multi-class classification-based segmentation, enabling accurate differential diagnosis of 14 common oral lesions-benign, pre-malignant, and malignant-across various mouth locations using photographic images. A dataset of 2,072 clinical images was used to train and validate the model. The proposed framework integrates EfficientNet-B3 for classification and ResNet-101-based Mask R-CNN for segmentation, achieving a classification accuracy of 74.49% and segmentation performance with an average precision (AP50) of 72.18. The gradient-weighted class activation map technique was applied to the model outputs to enable visualization of the specific areas that were most influential for predictive decisions made by the model. This significantly improves upon the state-of-the-art, where previous models achieved lower segmentation accuracy (AP50 < 50%). The framework not only classifies the lesion type but also delineates the lesion boundaries with high precision, which is critical for early detection and differential diagnosis in clinical practice.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".