Design of a reduced carry chain propagation adder using FinFET

Biomedicine is still attempting to overcome one of the profession's most
pressing problems: detecting brain tumors. Early detection of brain cancer is possible
with advanced technology or instruments. Classifying brain cancer types utilizing patent
brain images allows for automation in automated operations. Furthermore, the proposed
new method is used to distinguish between brain tumors and other brain illnesses. To
distinguish the cancer from the other parts of the brain, the input image is first preprocessed. Following that, the images are separated into different hues and levels before
being processed using the Grey Level Co-Occurrence and SURF extraction methods to
reveal crucial information in the photos. Genetic optimization reduces the size of the
retrieved attributes. An advanced learning technique is utilized to train and validate
tumor categorization based on cut-down characteristics. The technique's accuracy, error,
sensitivity, and specificity are all compared to the present method. The approach has a
90%+ accuracy rate, with less than 2% inaccuracy for all types of cancer. Finally, the
specificity and sensitivity are greater than 89% and 91%, respectively. Genetic
algorithms are more efficient because the methods used are more accurate and
specialized than the other ways.