Development of a COVID-19 Patients’ Fatality Prediction System Using Swarm Intelligent Convolution Neural Network

Aims: This work aims to develop a system that can be used accurately and timely predict the fatality of positively tested COVID-19 patient through use deep learning technique – swarm intelligent convolutional neural network.
 Methodology: The dataset in this study was acquired from Kaggle repository database. contains Lung Chest X-Ray images patients. were pre- processed obtain desired image quality for further processing. followed by segmenting pre-processed images. An Enhanced Firefly Algorithm (EFA) formulated applying roulette wheel selection procedure model movement process firefly as deterministic assist standard (FA) application Chaotic Sinusoidal Map Function attractive which establishes balance between exploration exploitation FA. EFA applied optimize Convolution Neural Network (CNN) hyper-parameters (number layers, number filters per layer, filter size batch size). segmented result subsequently presented EFA-CNN feature extraction prediction cases. models (EFA-CNN CNN) implemented using Matrix Laboratory 2020a software. evaluated specificity, sensitivity, false positive rate, accuracy, recognition time/rate determine performance developed models.
 Results: findings revealed performs better patients’ compared CNN model. It also discovered select optimal values architecture accounted improved accuracy reduced time Patients’ Fatality Prediction System.
 Conclusion: will both government healthcare workers providing needed computational capability level patient.