Performance Evaluation of a Classification Model for Oral Tumor Diagnosis
Bibliographic record
Abstract
This paper extracted features from region of interest of histopathology images, formulated a classification model for diagnosis, simulated the model and evaluated the performance of the model. This is with a view to developing a histopathology image classification model for oral tumor diagnosis. The input for the classification is the oral histopathology images obtained from Obafemi Awolowo University Dental Clinic histopathology archive. The model for oral tumor diagnosis was formulated using the multilayered perceptron type of artificial neural network. Image preprocessing on the images was done using Contrast Limited Adaptive Histogram Equalization (CLAHE), features were extracted using Gray Level Confusion Matrix (GLCM). The important features were identified using Sequential Forward Selection (SFS) algorithm. The model classified oral tumor diagnosis into tive classes: Ameloblastoma, Giant Cell Lesions, Pleomorphic Adenoma, Mucoepidermoid Carcinoma and Squamous Cell Carcinoma. The performance of the model was evaluated using specificity and sensitivity. The result obtained showed that the model yielded an average accuracy of 82.14%. The sensitivity and the specificity values of Ameloblastoma were 85.71% and 89.4%, of Giant Cell Lesions were 83.33% and 94.74%, of Pleomorphic Adenoma were 75% and 95.24%, of Mucoepidermoid Carcinoma were 100% and 100%, and of Squamous Cell Carcinoma were 71.43% and 94.74% respectively. The model is capable of assisting pathologists in making consistent and accurate diagnosis. It can be considered as a second opinion to augment a pathologist’s diagnostic decision.
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 | 0.003 | 0.006 |
| 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 | 0.000 |
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 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".