Rasheed Mehwish, Iqbal Muhammad Waseem, Jaffar Arfan, Ashraf Muhammad Usman, Almarhabi Khalid Ali, Alghamdi Ahmed Mohammed, Bahaddad Adel A
Department of Computer Science, Superior University, Lahore 54000, Pakistan.
Department of Software Engineering, Superior University, Lahore 54000, Pakistan.
Diagnostics (Basel). 2023 Apr 17;13(8):1451. doi: 10.3390/diagnostics13081451.
The human brain, primarily composed of white blood cells, is centered on the neurological system. Incorrectly positioned cells in the immune system, blood vessels, endocrine, glial, axon, and other cancer-causing tissues, can assemble to create a brain tumor. It is currently impossible to find cancer physically and make a diagnosis. The tumor can be found and recognized using the MRI-programmed division method. It takes a powerful segmentation technique to produce accurate output. This study examines a brain MRI scan and uses a technique to obtain a more precise image of the tumor-affected area. The critical aspects of the proposed method are the utilization of noisy MRI brain images, anisotropic noise removal filtering, segmentation with an SVM classifier, and isolation of the adjacent region from the normal morphological processes. Accurate brain MRI imaging is the primary goal of this strategy. The divided section of the cancer is placed on the actual image of a particular culture, but that is by no means the last step. The tumor is located by categorizing the pixel brightness in the filtered image. According to test findings, the SVM could partition data with 98% accuracy.
人类大脑主要由白细胞组成，以神经系统为中心。免疫系统、血管、内分泌、神经胶质、轴突和其他致癌组织中位置不当的细胞会聚集形成脑瘤。目前无法通过物理方式找到癌症并进行诊断。可以使用MRI编程分割方法来发现和识别肿瘤。需要强大的分割技术才能产生准确的输出。本研究对脑部MRI扫描进行了检查，并使用一种技术来获得肿瘤影响区域更精确的图像。所提出方法的关键方面包括利用有噪声的MRI脑部图像、各向异性噪声去除滤波、使用支持向量机分类器进行分割以及将相邻区域与正常形态过程隔离开来。准确的脑部MRI成像是该策略的主要目标。癌症的分割部分被放置在特定培养物的实际图像上，但这绝不是最后一步。通过对滤波后图像中的像素亮度进行分类来定位肿瘤。根据测试结果，支持向量机能够以98%的准确率对数据进行分割。
