An improved deep convolutional neural network by using hybrid optimization algorithms to detect and classify brain tumor using augmented MRI images

Automated brain tumor detection is becoming a highly considerable medical diagnosis research. In recent diagnoses, and classification are considered to employ machine learning deep techniques. Nevertheless, the accuracy performance of current models need be improved for suitable treatments. this paper, an improvement in convolutional ensured by adopting enhanced optimization algorithms, Thus, Deep Convolutional Neural Network (DCNN) based on Harris Hawks Optimization (HHO), called G-HHO has been considered. This hybridization features Grey Wolf (GWO) HHO give better results, limiting convergence rate enhancing performance. Moreover, Otsu thresholding adopted segment portion that emphasizes detection. Experimental studies conducted validate suggested method total number 2073 augmented MRI images. The technique’s was comparing it with nine existing algorithms huge images terms accuracy, precision, recall, f-measure, execution time, memory usage. comparison shows DCNN-G-HHO much more successful than methods, especially scoring 97%. Additionally, statistical analysis indicates approach faster utilizes less at identifying categorizing cancers MR implementation validation Python platform. relevant codes proposed available at: https://github.com/bryarahassan/DCNN-G-HHO .