CNOL: An Efficient Red Blood Cell Smears Based Malaria Disease Detection Using Digital Image Processing Technique with Convoluted Neural Optimization Logic

Malaria is a life-threatening disease affecting millions worldwide, and early diagnosis is critical for effective treatment. This study proposes CNOL (Convoluted Neural Optimization Logic), an efficient system for detecting malaria from red blood cell (RBC) smear images using digital image processing and neural network optimization. The CNOL model employs a convolutional neural network (CNN) architecture optimized through an advanced neural logic framework, enhancing its ability to classify infected and non-infected RBCs. A diverse dataset of blood smear images is processed through techniques such as image normalization, segmentation, and feature extraction to prepare data for model training. The model achieved a remarkable accuracy of 97.2%, with high precision (95.8%) and recall (96.9%), outperforming several existing methods like traditional CNNs, SVMs, and KNNs. The CNOL model demonstrated robust performance across all evaluation metrics, including a high AUC-ROC score of 98.1%, indicating its strong discriminatory ability. The model's time and resource efficiency make it suitable for real-time clinical use, requiring minimal computational power while providing accurate and timely diagnoses.