A new study shows that a multi model artificial intelligence (AI) framework combining deep learning and machine learning achieved an accuracy of 96.47% in malaria diagnosis, providing an efficient diagnostic tool for global malaria prevention and control. This research, jointly conducted by institutions such as Al Jawf University in Saudi Arabia and Cairo University in Egypt, has promoted the application of AI in the field of tropical disease diagnosis. Malaria, as a major health threat in tropical and subtropical regions, reported 249 million cases globally in 2022, with 90% of cases in Africa. Traditional microscopy relies on expert experience, is time-consuming and prone to errors, and is difficult to popularize in resource scarce areas. The multi model AI framework developed this time innovatively integrates three deep learning models for feature extraction. The team verified the excellent performance of the framework on the public dataset containing 27558 blood smear images: the sensitivity reached 96.03%, the specificity reached 96.90%, and the accuracy reached 96.88%, all superior to the single model. The majority voting mechanism effectively reduces the risk of misjudgment by integrating multiple models, and its performance surpasses existing AI diagnostic methods. This achievement provides a reliable solution for rapid diagnosis of malaria, and the technology can be deployed without special equipment, especially in areas with limited medical resources. Next, the team will expand dataset validation and explore applications in the diagnosis of other blood diseases such as leukemia. (New Society)