Brain tumor is a disorder caused by the growth of abnormal brain cells. Brain tumor is a major risk for the patient’s survival rate and quality of life since it can cause severe impairment of organ function and even death. In the study, the research team presented an approach by utilizing methods to the modeling of brain tumor by replacing 3D with 2D convolutions based on tumor segmentation on 2D images using the Geometric convolutional neural network (gCNN) and Point Cloud. This method not only enables to conduct an accurate reconstruction and location, but also aids in monitoring changes in tumor size and growth rate over time. A result of the study indicates that the implement of gCNN and Point Cloud significantly improves the accuracy and efficiency in brain tumor segmentation and monitoring, thus supporting medical practitioners in the diagnosis and treatment planning process. This approach holds immeasurable promise in the medical field, aiming to enhance the quality of care and treatment for patients with brain tumors.