Paper Title
Detection of Pneumonia using Multi Scale Convolution Mantaray System

Abstract
As of late, the explores and engineers from one side of the planet to the other are focussing on profound learning and picture handling techniques to revive the pneumonia analysis as those methodologies are equipped for handling various X-beam and figured tomography (CT) pictures. Profound learning approaches are shown in the clinical field for the prosperous investigation. In this review, crossover multiscale convolutional mantaray model is proposed for the recognition of pneumonia. The proposed model goes under three classifications specifically pre-handling, profound component extraction and grouping. The info dataset is rebuilt with the purpose of cross breed fluffy shaded and stacking approach. The freed of fluffy shading procedure is to destroy the spontaneous clamors and following that the hued picture get stacked with the first contribution by stacking way to deal with accomplish a best quality dataset. Then, at that point, the profound element extraction stage is handled by half and half multiscale highlight extraction unit to extricate numerous elements and furthermore the elements and organization size are decreased by the self-consideration module (SAM) based Convolutional Network (CNN). What's more, the blunder in the proposed network model is decreased with the guide of Adaptive Mantaray Foraging Optimization (AMRFO) approach. At last, the Support Vector Regression (SVR) is recommended to characterize the presence of pneumonia. The proposed strategy is contrasted and existing methodologies as far as exactness, accuracy, review, and F1-score of 97%, 95%, 97% and 96% individually. Catchphrases: Convolutional brain organization (CNN), Support vector relapse (SVR), Self-consideration module (SAM), Mantaray searching advancement, pneumonia, fluffy shading, stacking. Keywords - Convolutional Neural Network (CNN), Support Vector Regression (SVR), Self-attention Module (SAM), Mantaray Foraging Optimization, Pneumonia, Fuzzy Color, Stacking