Enactment of KNN in Brain Tumor Recognition: A Censorious Explication

• Main Authors: Hussein, A. H.A., Al-Hameed, Mazin Riyadh, Al-luhiby, Hussein A., Mohammed, E. Ali, Shakir, Aboothar mahmood, Saleem, Munqith
• Format: Conference Proceeding
• Language: English
• Subjects: Brain Tumor; C-means; Classification algorithms; Filtering algorithms; K-NN; Magnetic resonance imaging; Morphology; Recognition; Reliability; Surveillance; Task analysis; Technology
• ISSN: 2831-753X
• DOI: 10.1109/IICETA57613.2023.10351255

The accurate study of a brain tumor's morphology is truly a challenging process, and as a result, a computerized method for tumor detection is currently still being used. Unquestionably, this is more efficient than surveillance study and produces findings that are more reliable. The suggested strategy is innovative for both tumor identification and the capacity to determine the percentage of total brain cells that the tumor occupies. Brain tumors are characterized by the collection of aberrant cells in particular brain tissues. The patient's medication and prognosis are greatly influenced by the patient's past differentiating evidence of brain tumors. Finding and analyzing a brain tumor is frequently a difficult and time-consuming task. In this study, we have introduced a knowledge-based MRI brain tumor detection method for efficient categorization and grading of brain tumor images. Pre-screening, edge detection, identification, and fragmentation are the four components that make up the proposed framework. It is initially necessary to remove any noise from the source images using the Median Filter since noise could affect how accurately the identification process works. The images are instantly converted into three-dimensional blocks.