The presence of Plasmodium parasites causes malaria, which endangers human lives as it is a dangerous disease. Most of the existing work is computationally expensive or is not optimized for detecting miniature parasites and is slow at evaluating results. For prevailing the above-stated problems, we developed a hybrid diagnostic model that uses both Machine Learning and Deep Learning algorithms. This hybrid model uses a Convolutional Neural Network in combination with a Random Forest, where the features of the microscopic images are extracted with the aid of a CNN model. Using these features, we classify Malaria parasite images into parasitized and uninfected cells using a Random Forest algorithm. Over the past decade, models such as R-CNN, VGG-19, SVM, YOLOv4, and IDTL-MPDC have been evaluated and compared. Using a hybrid CNN model, we addressed the problems associated with previous work by splitting the dataset into 20% for testing, achieving 100% training accuracy and 96.04% testing accuracy.