Page 1 of 19

Archives of Business Research – Vol. 13, No. 03

Publication Date: March 25, 2025

DOI:10.14738/abr.1303.18433.

Coutinho, G. S., Medeiros, E. C., Fônseca, J. C. B., Moser, P. C., de Andrade, R. C. D., de Carvalho, F. F., de Souza Leão Júnior, F. P.,

& de Oliveira Domingues, M. A. (2025). Classification of Malaria-Infected Cells Using Convolutional Neural Networks: An Image- Based Microscopic Approach to Aid Diagnosis. Archives of Business Research, 13(03). 152-170.

Services for Science and Education – United Kingdom

Classification of Malaria-Infected Cells Using Convolutional

Neural Networks: An Image-Based Microscopic Approach to Aid

Diagnosis

Guilherme Silveira Coutinho

University of Pernambuco, Caruaru-PE, Brazil

Erika Carlos Medeiros

ORCID: 0000-0003-2506-7116

University of Pernambuco, Caruaru-PE, Brazil

Jorge Cavalcanti Barbosa Fônseca

University of Pernambuco, Caruaru-PE, Brazil

Patrícia Cristina Moser

University of Pernambuco, Caruaru-PE, Brazil

Rômulo César Dias de Andrade

University of Pernambuco, Caruaru-PE, Brazil

Fernando Ferreira de Carvalho

University of Pernambuco, Caruaru-PE, Brazil

Fernando Pontual de Souza Leão Júnior

University of Pernambuco, Caruaru-PE, Brazil

Marco Antônio de Oliveira Domingues

Federal Institute of Science and Technology of Pernambuco, Recife-PE, Brazil

ABSTRACT

This study investigates the use of convolutional neural networks to automatically

detect malaria-infected cells in blood smear images, offering an alternative to

manual diagnosis, which depends on specialized professionals and adequate

infrastructure. Manual diagnosis is time-consuming and prone to human errors,

especially in malaria-endemic regions with limited resources. Automated

approaches based on convolutional neural networks provide a promising solution

to optimize the diagnostic process and improve access to rapid treatment in remote

areas. The research evaluates the performance of different convolutional neural

network architectures for malaria diagnosis, following the Cross Industry Standard

Process for Data Mining methodology to structure preprocessing, modeling, and

model evaluation. Preprocessing involved normalization and data augmentation

techniques to enhance sample quality and diversity. Two architectures were

compared: a customized convolutional neural network designed to balance

computational efficiency and accuracy, and an adapted VGG16, recognized for its

Page 2 of 19

153

Coutinho, G. S., Medeiros, E. C., Fônseca, J. C. B., Moser, P. C., de Andrade, R. C. D., de Carvalho, F. F., de Souza Leão Júnior, F. P., & de Oliveira

Domingues, M. A. (2025). Classification of Malaria-Infected Cells Using Convolutional Neural Networks: An Image-Based Microscopic Approach to Aid

Diagnosis. Archives of Business Research, 13(03). 152-170.

URL: http://doi.org/10.14738/abr.1303.18433

advanced image feature extraction capabilities. Both models were trained and

evaluated using a robust dataset of blood cell images. The customized network

achieved 95.6% accuracy, outperforming similar models in the literature and

demonstrating its practicality for low-resource settings. It also exhibited high

precision, recall, and F1-score, ensuring reliable and balanced classification of

infected and healthy cells. This approach reduces reliance on specialists and

advanced equipment, making malaria diagnosis more accessible and efficient. The

findings position the customized convolutional neural network as a viable solution

for automated malaria diagnosis, combining simplicity with high performance and

offering significant potential for improving healthcare in endemic regions.

Keywords: Malaria Diagnosis, Convolutional Neural Network, Machine Learning, Image

Processing

INTRODUCTION

Malaria remains a significant public health threat, especially in tropical and subtropical regions,

where its high incidence and mortality rates directly impact the quality of life of affected

populations. According to the World Health Organization (WHO) report of 2023, the disease

affects millions of people every year, with more than 247 million cases reported in 2021 and an

average of 619,000 annual deaths. In Brazil, most cases occur in the Amazon region, which

accounts for approximately 99.9% of infections, reflecting the favorable environmental

conditions for the proliferation of the malaria vector, the Anopheles mosquito. This scenario

highlights the need for effective strategies for the diagnosis and control of the disease,

particularly considering the high transmissibility and risk of recurrence in endemic areas [1].

The early diagnosis of malaria is essential to interrupt its transmission and minimize its severe

consequences. According to the Malaria Treatment Guide in Brazil, the rapid identification of

Plasmodium, the parasite that causes the disease and is transmitted by the female Anopheles

mosquito, is crucial to prevent the worsening of symptoms and reduce the risk of severe

complications. The traditional diagnostic method, thick blood smear microscopy, is widely

recognized for its ability to detect and identify different species of Plasmodium. However, this

technique requires highly qualified professionals to ensure precision and reduce the margin of

error, which is challenging in areas with limited healthcare infrastructure. This scenario is even

more complex in remote and hard-to-reach regions, where the availability of equipment and

specialized professionals is restricted, underscoring the need to strengthen diagnostic

strategies for effective malaria control in Brazil [2].

Beyond diagnostic challenges, malaria poses a significant social and economic burden, deeply

affecting communities. In endemic regions, the disease undermines productivity, imposes high

costs on healthcare systems, and creates barriers to economic development. According to the

Epidemiological Bulletin published by the Brazilian Ministry of Health, without effective and

timely diagnosis, the ability to respond to malaria is compromised, which can intensify the

transmission cycle and increase the number of severe cases and deaths [3]. In this context,

implementing solutions that promote early infection identification is fundamental for

combating malaria, reducing mortality, and enabling more assertive interventions against the

disease [4]. Additionally, malaria negatively impacts the Human Development Index (HDI) of