@Article{dedt.2025.058943,
AUTHOR = {Khalid Jamil, Wahab Khan, Bilal Khan, Sarwar Shah Khan},
TITLE = {Advancing Brain Tumor Classification: Evaluating the Efficacy of Machine Learning Models Using Magnetic Resonance Imaging},
JOURNAL = {Digital Engineering and Digital Twin},
VOLUME = {3},
YEAR = {2025},
NUMBER = {1},
PAGES = {1--16},
URL = {http://www.techscience.com/dedt/v3n1/59694},
ISSN = {},
ABSTRACT = {Brain tumors are one of the deadliest cancers, partly because they’re often difficult to detect early or with precision. Standard Magnetic Resonance Imaging (MRI) imaging, though essential, has limitations, it can miss subtle or early-stage tumors, which delays diagnosis and affects patient outcomes. This study aims to tackle these challenges by exploring how machine learning (ML) can improve the accuracy of brain tumor identification from MRI scans. Motivated by the potential for artificial intillegence (AI) to boost diagnostic accuracy where traditional methods fall short, we tested several ML models, with a focus on the K-Nearest Neighbors (KNN) algorithm. Our results showed that KNN achieved impressive accuracy, reaching 94.09%, and outperformed other models in key metrics like the F1-Score and Matthews Correlation Coefficient (MCC). These findings underline the value of selecting the right AI model and suggest that KNN may be particularly useful in pinpointing tumors in MRI scans. This research highlights how AI-driven methods, particularly those that can incorporate richer imaging data, may offer a promising path to more reliable and early detection of brain tumors.},
DOI = {10.32604/dedt.2025.058943}
}