Automatic Detection of Ascaris Lumbricoides in Microscopic Images using Convolutional Neural Networks (CNN)
Giovanni Gelber Martinez Pastor, Cesar Roberto Ancco Ruelas, Eveling Castro-Gutiérrez, Victor Luis Vásquez Huerta · International Journal of Advanced Computer Science and Applications · 2024
Parasites are disease-causing agents both in Peru and worldwide. In many contexts, diagnosis is done manually by observing microscopic images, where it's necessary to identify parasite eggs. However, this process is notably slow, and sometimes image clarity may be insufficient, making rapid and accurate identification challenging. This can be due to various factors, such as image quality or the presence of noise. This paper focused on a Convolutional Neural Network (CNN) model. Through this approach, the training, testing, and validation stages of our CNN model to detect and identify Ascaris lumbricoides parasite eggs. The results show that the proposed CNN model, combined with image preprocessing, yielded highly favorable results in parasite egg identification. Additionally, very satisfactory values were achieved in model testing and validation, indicating its effectiveness and precision in diagnosing parasite presence. This research represents a significant advancement in the field of parasitological diagnosis, offering an efficient and accurate solution for parasite detection through microscopic image analysis. It is hoped that these results contribute to improving diagnosis and treatment methods for parasitic diseases.