GRAPH CONVOLUTIONAL NETWORKS FOR PREDICTING DISEASE OUTBREAKS IN PUBLIC HEALTH SURVEILLANCE

Abstract

Aim: With the help of CT scans, we construct a detection module by adhering to a procedure for COVID-19

Background: The current public health crisis, which is known as SARS-CoV-2, has resulted in a number of fatalities and has caused extensive economic disruption worldwide.

Methodology: By adhering to a pre-processing, feature-extraction, and detection strategy, we are able to construct a detection module that is capable of identifying COVID-19 patients via CT images. Following the extraction of features using a Grey Level Co-occurrence Matrix (GLCM), the next step in the image pre-processing process is classification using Graph Convolutional Networks (GCN).

Contribution:The objective of the simulation is to assess the performance of the model by making use of a number of different CT imaging datasets that contain images depicting a significant number of patients individually.

Findings: With a detection rate of 98% and a mean average percentage error (MAPE) that is lower than 0.2, the outcomes of the simulation reveal that the recommended method beats the traditional procedures that are currently in use.