Mpoxnet: Fast Detection of Monkeypox using Deep Learning in Digital Skin Images

Monkeypox is a communicable illness that causes various neurological disorders in addition to skin lesions, rashes, fever, respiratory difficulty, and lymph enlargement. This can be a fatal illness, and the most recent outbreak has demonstrated that it has expanded to Africa, the United States, Europe, and Australia. Usually, PCR is used to diagnose monkeypox once a sample of the skin lesion is obtained. A serious issue to scientists, researchers, medical professionals, governments, and the medical community globally, monkeypox has evolved from its discovery to its treatment. The researchers are working hard in their different fields to find any potential answers to stop this pandemic. The PCR test is one of the most popular and useful techniques used by researchers. One of the difficult jobs during a pandemic is manually reviewing every complaint, which takes time and multiple skin specialists. The proposed method uses deep learning algorithms to categorise skin lesions as monkeypox positive or negative. The suggested approach is evaluated using the Monkeypox Skin Image Dataset (MSID) that is made accessible on Kaggle with a 97.72% success rate by employing modified VGG16 i.e. Mpoxnet; these highly favourable findings suggest that it may be widely used to diagnose monkeypox. This technique can be applied successfully in impoverished places where laboratory infrastructure could be inadequate. In this work, a deep learning neural network-based technique is used (i.e mpox) presented as an alternate quick screening method for monkeypox that works by examining patient images to search for visual cues present in the skin condition imaging of monkeypox patients.