Dataset Information

An ensemble-based approach for automated medical diagnosis of malaria using EfficientNet.

ABSTRACT: Each year, more than 400,000 people die of malaria. Malaria is a mosquito-borne transmissible infection that affects humans and other animals. According to World Health Organization (WHO), 1.5 billion malaria cases and 7.6 million related deaths have been prevented from 2000 to 2019. Malaria is a disease that can be treated if early detected. We propose a support decision system for detecting malaria from microscopic peripheral blood cells images through convolutional neural networks (CNN). The proposed model is based on EfficientNetB0 architecture. The results are validated with 10-fold stratified cross-validation. This paper presents the classification findings using images from malaria patients and normal patients. The proposed approach is compared and outperforms the related work. Furthermore, the proposed ensemble method shows a recall value of 98.82%, a precision value of 97.74%, an F1-score of 98.28% and a ROC value of 99.76%. This work suggests that EfficientNet is a reliable architecture for automatic medical diagnostics of malaria.

Supplementary information

The online version contains supplementary material available at 10.1007/s11042-022-12624-6.

SUBMITTER: Marques G

PROVIDER: S-EPMC8964254 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

An ensemble-based approach for automated medical diagnosis of malaria using EfficientNet.

Marques Gonçalo G Ferreras Antonio A de la Torre-Diez Isabel I

Multimedia tools and applications 20220329 19

Each year, more than 400,000 people die of malaria. Malaria is a mosquito-borne transmissible infection that affects humans and other animals. According to World Health Organization (WHO), 1.5 billion malaria cases and 7.6 million related deaths have been prevented from 2000 to 2019. Malaria is a disease that can be treated if early detected. We propose a support decision system for detecting malaria from microscopic peripheral blood cells images through convolutional neural networks (CNN). The ...[more]

PMID: 35368860

Similar Datasets

Project description:BackgroundEarly and accurate diagnosis of malaria is a critical aspect of efforts to control the disease, and several diagnostic tools are available. Microscopic assessment of a peripheral blood smear enables direct visualization of parasites in infected red blood cells and is the clinical diagnostic gold standard. However, it is subjective and requires a high level of skill. Numerous indirect detection methods are in use, but are not ideal since surrogate markers of infection are measured. This study describes the first clinical performance evaluation of the automated Sysmex XN-30 analyser, which utilizes fluorescence flow cytometry to directly detect and quantitate parasite-infected red blood cells.ResultsResidual EDTA blood samples from suspected malaria cases referred for routine diagnosis were analysed on the XN-30. Parasitaemia was reported as a percentage, as well as absolute numbers of infected red blood cells, and scattergrams provided a visual image of the parasitized red blood cell clusters. The results reported by the XN-30 correlated with microscopy and the analyser demonstrated 100% sensitivity and specificity. Measurements were reproducible and storage of samples at room temperature did not affect the parameters. Several Plasmodium species were detected, including Plasmodium falciparum, Plasmodium vivax and Plasmodium ovale. The XN-30 also identified the transmissible gametocytes as a separate cluster on the scattergrams. Abnormal red blood cell indices (low haemoglobin and raised reticulocyte counts), haemoglobinopathies and thrombocytopenia did not interfere with the detection of parasites. The XN-30 also generated a concurrent full blood count for each sample.ConclusionsThe novel technology of the Sysmex XN-30 provides a robust, rapid, automated and accurate platform for diagnosing malaria in a clinical setting. The objective enumeration of red blood cells infected with Plasmodium species makes it suitable for global use and allows monitoring of the parasite load once therapy has been initiated, thereby providing an early marker of drug resistance. The automated generation of a full blood count for each sample provides an opportunity for detecting unsuspected cases. Asymptomatic carriers can also be identified, which will be useful in blood transfusion centres, and will enable treatment of these individuals to prevent the spread of the disease.

Dataset Information

An ensemble-based approach for automated medical diagnosis of malaria using EfficientNet.

Supplementary information

Publications

An ensemble-based approach for automated medical diagnosis of malaria using EfficientNet.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets