Abstract
Malaria is one of the most widespread and deadly diseases across the globe, especially in sub-Saharan Africa and other parts of the developing world. This is primarily because of incorrect or late diagnosis. Existing diagnostic techniques mainly depend on the microscopic identification of parasites in the blood smear stained with special dyes, which have drawbacks such as being time-consuming, depending on skilled personnel and being vulnerable to errors.
This work seeks to overcome these challenges by proposing a deep learning-based solution in the ConvNeXt architecture incorporating transfer learning and data augmentation to automate malaria parasite identification in thin blood smear images. This study’s dataset was a set of blood smear images of equal numbers of parasitised and uninfected samples drawn from a public database of malaria patients in Bangladesh. To detect malaria in the given dataset of parasitised and uninfected blood smears, the ConvNeXt models were fine-tuned. To improve the effectiveness of these models, a vast number of data augmentation strategies was used so that the models could work well in various image capture conditions and perform well even in environments with limited resources. The ConvNeXt Tiny model performed better, particularly the re-tuned version, than other models, such as Swin Tiny, ResNet18, and ResNet50, with an accuracy of 95%. On the other hand, the re-modified version of the ConvNeXt V2 Tiny model reached 98% accuracy. These findings show the potential to implement ConvNeXt-based systems in regions with scarce healthcare facilities for effective and affordable malaria diagnosis.
Competing Interest Statement
The authors have declared no competing interest.
Funding Statement
The author(s) received no specific funding for this work.
Author Declarations
I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.
Not Applicable
The details of the IRB/oversight body that provided approval or exemption for the research described are given below:
The Institutional Review Board (IRB) approved using the data at the NLM (IRB#12972)(11). Additionally, this study received ethical clearance from the University of Johannesburg’s School of Consumer Intelligence and Information Systems Research Ethics Committee (SCiiSREC) under ethical clearance code 2024SCiiS040
I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.
Not Applicable
I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).
Not Applicable
I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.
Not Applicable
Data Availability
All image files used in this study are available from https://lhncbc.nlm.nih.gov/LHC-research/LHC-projects/image-processing/malaria-datasheet.html
https://lhncbc.nlm.nih.gov/LHC-research/LHC-projects/image-processing/malaria-datasheet.html
