Monkeypox recognition and prediction from visuals using deep transfer learning-based neural networks

As the globe struggles to recover from COVID-19, the monkeypox virus has emerged as a new global pandemic threat. Monkeypox cases are still being reported daily from different nations despite the virus not being as harmful or contagious as COVID-19. As a result, the possibility of another worldwide pandemic occurring directly due to a lack of adequate preventative measures will not come as a complete shock to everyone. Diagnosing Monkeypox in its early stages may be challenging because it resembles chickenpox and measles. When confirmatory Polymerase Chain Reaction assays are not readily available, monitoring suspected cases and swiftly detecting them may be possible with computer-assisted detection of monkeypox lesions. Recent research has shown that deep learning models have significant promise for image-based diagnostics, including cancer diagnosis, identifying tumor cells, and detecting COVID-19 patients. To address these challenges, we built a deep learning model based on transfer learning that can assist medical professionals and other individuals in determining whether they are suffering from Monkeypox. The InceptionV3 model utilized in this study was trained with the publicly accessible Monkeypox dataset. During the studies, the model attained an accuracy of 98%.