Three Dimensional DenseUNet with CKHA Segmentation Technique for Monkeypox Disease Prediction

Since several years ago, the monkeypox virus has been seen in Africa, where it has been connected to the emergence of skin lesions. Following the COVID-19 pandemic, lethal consequences of viral infections have caused public concern and anxiety. Given that COVID-19 has reached pandemic levels, quick detection techniques are essential. The hybrid pre-trained deep learning (DL) models are used in this study's research to identify the monkeypox virus. This method uses the Messy Krill Guide Procedure (CKHA) in conjunction with Kapur's entropy (KE) to segment images. The Remora Optimisation Algorithm (ROA) selects the Long-Short Term Memory (LSTM) weight in the classifier in the best possible way. For this, we first adjust them by adding familiar routine sheets for all of them, and then we examine the outcomes via tetrad familiar-known metrics: Accuracy, Remember, F1notch, and Correctness. With the aid of our suggested segmentation and classification strategy, the study runs its trials on a publically accessible dataset, yielding average values of 96.90% accuracy, 97.84% accuracy, 98.20% remember, and 98.67% F1-notch, respectively. These optimistic outcomes, which exceed cutting-edge techniques, imply that the suggested strategy is appropriate for mass screening by health practitioners.