North South University Researcher Advances Knowledge in Oral Squamous Cell Carcinoma (Predicting Leukoplakia and Oral Squamous Cell Carcinoma Using Interpretable Machine Learning: A Retrospective Analysis).

A recent study conducted by researchers at North South University in Dhaka, Bangladesh, aimed to assess the effectiveness of interpretable machine learning models in diagnosing leukoplakia and oral squamous cell carcinoma (OSCC). The study analyzed data from 237 patient cases, including information on patient demographics, lesion characteristics, and lifestyle factors. The Random Forest model was found to have the highest accuracy (93%) and Kappa score (0.90) in diagnosing OSCC. The study also identified key features that influenced the prediction of leukoplakia and OSCC. The researchers concluded that the Random Forest model outperformed other machine learning models in diagnosing oral cancer and potentially malignant oral lesions, and the use of SHAP improved the understanding of the importance of features in facilitating early diagnosis. [Extracted from the article]