Research Article
Design of Malaria Detection Using Ensemble Techniques - A Combination of Alexnet and Densenet Algorithm
@INPROCEEDINGS{10.4108/eai.23-11-2023.2343229, author={Meena Preethi B and Gowri P and Karthikeyan C and Dharshini B and Parameshvar M and Gokul S}, title={Design of Malaria Detection Using Ensemble Techniques - A Combination of Alexnet and Densenet Algorithm}, proceedings={Proceedings of the 1st International Conference on Artificial Intelligence, Communication, IoT, Data Engineering and Security, IACIDS 2023, 23-25 November 2023, Lavasa, Pune, India}, publisher={EAI}, proceedings_a={IACIDS}, year={2024}, month={3}, keywords={malaria ensemble technique alexnet densenet deep learning}, doi={10.4108/eai.23-11-2023.2343229} }- Meena Preethi B
Gowri P
Karthikeyan C
Dharshini B
Parameshvar M
Gokul S
Year: 2024
Design of Malaria Detection Using Ensemble Techniques - A Combination of Alexnet and Densenet Algorithm
IACIDS
EAI
DOI: 10.4108/eai.23-11-2023.2343229
Abstract
This paper aims to identify whether a cell is malaria infected or not by applying machine learning and Deep learning algorithms that are Alex net and Dense net. The dataset which is used for reference consists of a total 27,558 images, out of which 13,780 images are infected, and the rest are uninfected cells and is taken from the NIH Website. In the paper, the sample of that dataset is taken, and algorithms are applied to evaluate the dataset. The machine is trained to classify and detect if the cell is parasitized or uninfected. An in-breadth and depth analysis of various classifiers like AlexNet and DenseNet is done, and their performance is compared by tuning different hyperparameters. DenseNet is a deep convolutional neural network architecture developed for image classification tasks. It is characterized by densely connected layers, which enables better feature reuse and gradient flow throughout the network. AlexNet is a deep neural network that can learn complex features from images andis very effective at image classification tasks. The output is combined using an ensemble technique, and the performance of these classifiers is evaluated.
