Abstract Tumors are difficult to notice in medical imaging due to their intricate structure and noise, making it difficult and time-consuming for doctors to locate them. This is important since locating and pinpointing the tumor’s site at an early stage is critical. Due to the complicated structure of tumors and the involution of noise in Magnetic Resonance Imaging (MRI) data, physical tumors identification has become a difficult and time-consuming process for medical practitioners. Thus, this paper proposes a machine learning-based approach for segmenting and categorizing of magnetic resonance images to identify brain tumors. The framework of this paper uses SVM and Naive Bayes algorithms for image pre-processing, feature extraction, and classification. The results indicated that the two classification algorithms used (Naïve Bayes and SVM) had an accuracy of 0.89 for SVM and 0.51 for naïve Bayes, a sensitivity of 0.57 and 0.85, and a specificity of 0.99 0.42, respectively. The findings indicate that SVM is more precise and specified than Naïve Bayes but that Naïve Bayes is more sensitive, with a sensitivity of 0.85. The Naïve Bayes classifier produces modest performance when compared to SVM classifiers.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|