What does this research mean for the field?

Computational morphometry combined with machine learning algorithms, particularly Random Forest, can objectively and accurately classify the severity of oral epithelial dysplasia based on nuclear and cellular features. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To evaluate whether automated computational morphometry combined with machine learning algorithms can provide an objective, reproducible histopathological risk stratification for oral epithelial dysplasia.

September 2, 2026Advances in Clinical Medical ResearchOpen Access

Computational Morphometric Analysis and Machine Learning-Based Classification of Oral Epithelial Dysplasia: An Objective Approach to Histopathological Risk Stratification

View Full Paper

Ask AI

Bookmark

Authors

MKM. Kuppusamy MGManish Gupta RSRajendra Singh

Discussion

Member takes

Implication

Histopathological study demonstrates machine learning accurately stratifies oral epithelial dysplasia using cellular morphometry, suggesting an objective adjunct to conventional grading.

Key Points

To evaluate whether automated computational morphometry combined with machine learning algorithms can provide an objective, reproducible histopathological risk stratification for oral epithelial dysplasia.
Analyzed digital photomicrographs of 81 oral tissue specimens, evenly divided across normal buccal mucosa (n=27), low-risk dysplasia (n=27), and high-risk dysplasia (n=27).
Extracted seven nuclear and cellular morphometric parameters using QuPath, including nuclear-to-cytoplasmic ratio, circularity, area, and perimeter.
Trained and evaluated six machine learning algorithms (Random Forest, Decision Tree, Support Vector Machine, Logistic Regression, K-Nearest Neighbors, and Gaussian Naïve Bayes) for classification performance.
Statistically significant morphometric alterations appeared across groups, characterized by progressive increases in nuclear area, nuclear perimeter, and nuclear-to-cytoplasmic ratio, alongside reductions in cellular area and nuclear circularity.
Random Forest demonstrated the highest overall classification accuracy at 86.73%, followed by Decision Tree (78.96%), Support Vector Machine (68.42%), Logistic Regression (64.38%), K-Nearest Neighbors (59.84%), and Gaussian Naïve Bayes (52.47%).

Cite This Study

Kuppusamy et al. (2026) studied this question.

synapsesocial.com/papers/6a97e2f6c562ede874ec773c https://doi.org/10.5281/zenodo.22192848

View Full Paper

Ask AI

Bookmark

Implication

Histopathological study demonstrates machine learning accurately stratifies oral epithelial dysplasia using cellular morphometry, suggesting an objective adjunct to conventional grading.

Computational Morphometric Analysis and Machine Learning-Based Classification of Oral Epithelial Dysplasia: An Objective Approach to Histopathological Risk Stratification

Discussion

Member takes

Implication

Key Points

Cite This Study

Computational Morphometric Analysis and Machine Learning-Based Classification of Oral Epithelial Dysplasia: An Objective Approach to Histopathological Risk Stratification

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider