An Adaptive Convolutional Neural Network-Random Forest Model for Human Monkeypox Detection and Severity Grading

The zoonotic viral illness known as monkeypox poses a substantial threat to public health because it can be transmitted from human to human and can emerge in a wide variety of clinical presentations. Human-to-human transmission and a variety of clinical manifestations make monkeypox a major public health issue. Clinical care and epidemic containment depend on early and accurate diagnosis and disease severity classification. The authors present a groundbreaking hybrid model that seamlessly combines the power of Convolutional Neural Networks (CNNs) and the comprehensibility of Random Forests to boldly tackle monkeypox disease detection and disease severity classification. CNNs are used to extract complex traits from monkeypox-infected lesion pictures, maintaining spatial information. After sophisticated feature extraction, the model expertly classifies cases into severity categories using Random Forests. With this advanced method, they can assess illness development more precisely than with binary classifications. Well-curated datasets that cover the range of monkeypox cases have supported the model's development and assessment. This dataset is the foundation on which the model is refined and proven to meet real-world obstacles with unmatched precision and reliability. The model's accuracy rate is 79.03%, making it exceptional. Notably, it can distinguish between ten severity categories, exhibiting unmatched clinical acuity. This study advances monkeypox detection and severity grading. Clinician decision-making, public health response tactics, and communities struggling with monkeypox will benefit from its revolutionary impact.