Brain tumor classifications by gradient and XG boosting machine learning models

Brain tumor (BT) is the mass or development of uncommon cells in an individual brain. Various kinds of cerebrum tumors exist with dangerous (Malignant) and nondangerous (Benignant). However, health practitioners are recommending implementing machine learning (ML) algorithms in large medical image datasets and identifying the patterns to promote the decision making of health experts. Therefore, we incorporated the well-known boosting algorithms known as Gradient boosting and XG boosting for the classification of BT images. Model validation was performed by cross-validation (k-fold) techniques and performance metrics were presented in confusion matrix and accuracy terms. With Gradient boosting, we have achieved 96.6% of accuracy, and then this value has been further improved by XG boosting technique and achieved 97.2% of accuracy.