PERBANDINGAN METODE DECISION TREE DENGAN NAÃVE BAYES DALAM KLASIFIKASI TUMOR OTAK CITRA MRI
Total View This Week0
Total View This Week0
Institusion
Universitas Pembangunan Nasional Veteran Jakarta
Author
Suci Dilasari Kamil, .
Subject
T Technology (General)
Datestamp
2022-01-12 04:53:16
Abstract :
In medical image classification, Machine Learning algorithm is commonly implemented. Decision Tree and Naive Bayes are commonly used method in medical image classification. Therefore, a comparison between Decision Tree and Naive Bayes algorithm is concluded to get the performance of the classification methods to MRI, with preprocess of grayscale, K- means clustering for segmentation, and GLCM for texture feature extraction. This study will implement texture analysis with contrast, correlation, energy, and homogeneity to classify the images to two class: brain tumor and non-brain tumor. From the study, based on the value of accuracy, specificity, and sensitivity, Decision Tree has higher values compared to Naive Bayes which are 96% accuracy, 96% specificity, and 96% sensitivity compared to Naive Bayes value of 91% accuracy, 90% specificity, and 93% sensitivity.
In medical image classification, Machine Learning algorithm is commonly implemented. Decision Tree and Naive Bayes are commonly used method in medical image classification. Therefore, a comparison between Decision Tree and Naive Bayes algorithm is concluded to get the performance of the classification methods to MRI, with preprocess of grayscale, K- means clustering for segmentation, and GLCM for texture feature extraction. This study will implement texture analysis with contrast, correlation, energy, and homogeneity to classify the images to two class: brain tumor and non-brain tumor. From the study, based on the value of accuracy, specificity, and sensitivity, Decision Tree has higher values compared to Naive Bayes which are 96% accuracy, 96% specificity, and 96% sensitivity compared to Naive Bayes value of 91% accuracy, 90% specificity, and 93% sensitivity.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
