Neural Network based approach to diagnose and classify Monkeypox disease

Abstract

Numerous organizations, including local governments, health, and medical institutions, and even the WHO, have expressed concern about the fatality and transmission rates of this illness due to the rapid increase in monkeypox cases around the world. Monkeypox illness and the COVID-19 virus have a very similar pattern of spread. The number of reported cases worldwide has increased in recent months, continuing the same trend as COVID-19. The characteristics and signs of the monkeypox virus are similar to those of any other viral illness. The main signs and symptoms are fever, chills, fatigue, headache, etc. At this juncture, the diagnosis is a significant challenge. Another challenge is posed by the disease when the obvious symptoms occur like rashes on the skin. The problem with these specific symptoms is that their appearance resembles other diseases like Chickenpox, Cowpox, and so on. Thus, the correct classification and proper diagnosis of the disease are tough and strenuous. Thus, for efficient and correct classification of this severe epidemic, a neural network-based solution is proposed, which classifies the disease at its initial stage with a competent accuracy rate of 94.38%. the proposed solution is excelling the diagnosis problem by performing efficient classification.