Detection of malaria disease using novel machine learning-oriented optimization concept

Malaria still poses a serious threat to world health, particularly in tropical as well as subtropical regions, where it kills hundreds of thousands of people every year. Effective management as well as intervention depend on accurate and timely identification. However, there exists several barriers to current diagnostic techniques, including limited access to healthcare, a shortage of skilled professionals, as well as the inconsistent accuracy of traditional techniques such as microscopy as well as quick diagnostic testing. These limitations highlight the urgent need for innovative, scalable strategies to improve early diagnosis as well as decrease the burden of malaria globally. Hence, this paper accomplishes the malaria disease detection using novel machine learning-based optimization methodology. From the online reputed sources, the dataset is initially gathered. The pre-processing of this gathered data is performed by the median filter approach. The features are next extracted from these pre-processed data by the contour features method. Finally, the novel Improved Extreme Learning Machine (IELM) model does the detection of the malaria disease. The parameter tuning in ELM takes place by nature inspired optimization algorithm called Reindeer Cyclone Optimization Algorithm (RCOA) with accuracy maximization as the fitness function. This research will be a useful tool for illness identification and employs a novel approach for quick processing. The proposed IELM-RCOA for the malaria disease detection model is 5.62 and 6.01% better than the other existing models considered in terms of accuracy and recall, respectively.