MalariaMalaria is an epidemic disease which causes by parasites that are spread through bites from infected Anopheles mosquitoes. Through early diagnosisDiagnosis and timely treatment, malariaMalaria death rates can be decreased and prevented. But the manual examination of blood smears by laboratory technicians is time-consuming and probable chances of human errors. Considering this need, in this paper, we are demonstrating a proposed system deployed over a WebApp could help laboratories to perform fast and accurate tests. For the malariaMalaria parasite detection in the blood cells, we proposed our custom CNN modelCustom CNN model and even implemented the dataset on the Transfer Learning model (VGG19) to compare the results. We got the highest accuracy of 97.74% for our CNN model which is the simplest of the current models as it prioritizes only important parameters, provides great computational efficiency, and is realistic for implementation. Additionally, a malariaMalaria outbreak warning system is proposed to warn the people in localities when there is availing risk based on climatic factors. A logistic regressionLogistic Regression model implemented by Gradient Descent Optimizer is deployed which predicts the malariaMalaria outbreak from the live weather forecastForecast inputs from the OpenWeather API and displays the probability of the outbreak to the users with one click.
Malaria Parasite Detection and Outbreak Warning System Using Deep Learning
Lect. Notes Electrical Eng.
International Conference on Robotics, Control, Automation and Artificial Intelligence ; 2022 November 24, 2022 - November 26, 2022
2023-11-18
16 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Parasite detection and identification for automated thin blood film malaria diagnosis
British Library Online Contents | 2010
|A Forward Collision Warning System Using Deep Reinforcement Learning
SAE Technical Papers | 2020
|Europäisches Patentamt | 2023
|