Paper Title
IMPROVING THE DETECTION OF KIDNEY TUMOURS USING FIREFLY ALGORITHM OPTIMISATION

Abstract
Rapid and accurate diagnosis of kidney tumors is essential for successful treatment and timely diagnosis. This work aims to provide new methods for better segmentation of cognitive tumors using deep learning techniques and firefly algorithm (FA). Cancer detection systems can be improved by using FA based on the flame behavior. Use this FA to fine tune feature selection and important parameters. Improving medical images is the first step of our plan. Then, feature extraction is performed using convolutional neural networks (CNNs) to capture important features indicative of kidney tumors. When FA optimizes the CNN hyperparameter and feature selection, it provides better detection accuracy and enables more efficient estimation. Experiments conducted on a publicly available dataset of psychotic tumors show significant improvements in classification accuracy, recall, precision, and F1 scores compared to traditional methods Clinicians benefit from the FA-optimized model from that the results show how well it consistently detects and classifies kidney tumors. Improvements in diagnostic accuracy and patient outcomes are possible thanks to the combination of bio-inspired optimization algorithms and modern deep learning techniques to solve complex medical imaging problems. Keywords - Kidney tumor detection, firefly algorithm, CNN, Hyperparameter tuning, image preprocessing.