Artificial intelligence (AI) has demonstrated significant promise in medical diagnostics. Healthcare providers can enhance patient outcomes and advance personalized healthcare systems by utilizing these AI-driven tools to make quicker, more accurate diagnoses. Despite its capabilities, AI has not yet achieved the same level of human intelligence. The main contribution of this paper is to present a new strategy called Adaptive Disease Diagnosis Strategy (ADDS). ADDS introduces a new approach to disease diagnosis, particularly for emergent diseases such as monkeypox virus (MPXV), by imitating human learning processes in contrast to conventional diagnostic procedures that depend on static models. ADDS uses the binary groupers and moray eels (BGME) optimization algorithm to select the most critical features from the MPXV dataset. ADDS is based on a modified version of Naive Bayes (NB), which is called Enhanced Incremental NB (EINB). EINB emulates how humans acquire knowledge by continually adapting to new information while building upon prior knowledge. It modifies its diagnostic capabilities without necessitating complete retraining. EINB optimizes the model's efficacy by adding a selective data approach that only incorporates the most significant information into the learning process, thereby avoiding useless updates from irrelevant data. This selective approach enables the model to retain previous information, guaranteeing that past data is valuable as new insights are incorporated into the model. The results demonstrate that the ADDS significantly enhances the performance of the monkeypox diagnostic system with an accuracy value equal to 99.46%, ensuring that the model remains accurate, adaptive, and ready for emerging challenges.