A Novel Framework for Malaria Detection Using Machine Learning
Ponam Roshitha, Ravuri Abhinaya, Pasala Bharatha Simha Reddy, Pasupuleti Reddy Kumar, C. Nalini · 2025
This research work proposes a novel approach for malaria diagnosis employing a Lightweight Convolutional Neural Network. The process involves preprocessing steps such as noise removal and image restoration through means clustering-based segmentation. Subsequently, a custombuilt Lightweight CNN model is introduced, integrating various convolutional layers tailored to the specific task. Leveraging deep learning algorithms, the model conducts classification to determine whether the input image depicts an infected or uninfected sample. The key focus lies in achieving high accuracy rates in malaria detection. The proposed technique's effectiveness is assessed thru complete trying out and experimentation. By integrating the lightweight architecture of the CNN model with superior picture processing strategies, the technique strives to enhance diagnostic accuracy and reduce the prevalence of fake positives and fake negatives. This research contributes to the ongoing efforts in medical image analysis, particularly in combating infectious diseases like malaria, by providing a robust and efficient framework for automated diagnosis.