Rancang Bangun Aplikasi Berbasis Web Untuk Sistem Deteksi Penyakit Malaria Menggunakan Computer Vision

Abstract

This study discusses the design and development of a web-based application for malaria
disease detection using Computer Vision. The main objective of this research is to develop
a system that can assist medical personnel in detecting malaria infections more quickly and
accurately by utilizing computer vision technology. The proposed system employs the
Convolutional Neural Network (CNN) algorithm to analyze microscopic erythrocyte
images obtained from blood samples. Malaria image data used in this study were collected
from various online sources as well as through direct observation in hospitals.The system
development process begins with the collection of malaria image data, which is then
processed through preprocessing stages to enhance data quality and prepare it for model
training. Once the data is ready, the CNN model is trained using augmented training data
to improve the model's generalization. Model evaluation is conducted using test data to
measure the accuracy and performance of the model in detecting malaria.Evaluation results indicate that the developed CNN model has high accuracy in detecting malaria infections,
with satisfactory precision, recall, and F1-score values. The system is also capable of
generating detection reports and visualizations that facilitate medical personnel in
diagnosis. This system is expected to support efforts to improve malaria diagnosis in
healthcare facilities, particularly in remote areas with limited access to trained medical
personnel. Additionally, the system is anticipated to reduce the workload of medical
personnel and increase efficiency in the malaria diagnosis process.