Classification for avian malaria parasite Plasmodium gallinaceum blood stages by using deep convolutional neural networks

Abstract

The infection of an avian malaria parasite (Plasmodium gallinaceum) in domestic chickens presents a major threat to the poultry industry because it causes economic loss in both the quality and quantity of meat and egg production. Computer-aided diagnosis has been developed to automatically identify avian malaria infections and classify the blood infection stage development. In this study, four types of deep convolutional neural networks, namely Darknet, Darknet19, Darknet19-448 and Densenet201 are used to classify P. gallinaceum blood stages. We randomly collected a dataset of 12,761 single-cell images consisting of three parasite stages from ten-infected blood films stained by Giemsa. All images were confirmed by three well-trained examiners. The study mainly compared several image classification models and used both qualitative and quantitative data for the evaluation of the proposed models. In the model-wise comparison, the four neural network models gave us high values with a mean average accuracy of at least 97%. The Darknet can reproduce a superior performance in the classification of the P. gallinaceum development stages across any other model architectures. Furthermore, the Darknet has the best performance in multiple class-wise classification, with average values of greater than 99% in accuracy, specificity, and sensitivity. It also has a low misclassification rate (< 1%) than the other three models. Therefore, the model is more suitable in the classification of P. gallinaceum blood stages. The findings could help us create a fast-screening method to help non-experts in field studies where there is a lack of specialized instruments for avian malaria diagnostics.

A novel approach for identification of zoonotic trypanosome utilizing deep metric learning and vector database-based image retrieval system

“…The application of YOLOv4-tiny-based object detection is proposed for cropping a single blood stage parasite from those present in a microscopic image. Among various YOLO algorithms, the YOLOv4-tiny model outperforms others in terms of accurate localization and classification [ 8 , 31 ], as demonstrated by its high balance accuracy [ [32] , [33] , [34] ]. Specifically, the inference capability of this model reaches up to 155 frames per second during processing.…”

“…The performance of all proposed models will be assessed using various statistical metrics, including a confusion matrix table, precision, recall, accuracy, F1-Score, and specificity [ 8 ]. These evaluation metrics were obtained and analyzed from the confusion matrix tables, and the formulas for these metrics are as described in equations (5) , (6) , (7) , (8) , (9) as follows: where TP represents the total number of true positives, TN is the total number of true negatives, FP denotes the total number of false positives, and FN is for the total number of false negatives classifications.…”

“…The observation of pathological changes at the microscopic level can significantly contribute to clinical decision-making. At present, machine learning (ML) and its subfield of algorithms, deep learning (DL), are employed for the diagnosis of malaria infection through the analysis of microscopic images [ [6] , [7] , [8] ]. Several neural network models, including the Single-Shot MultiBox Detector (SSD) model [ 6 ], adaptive color segmentation and tree-based decision classification [ 9 ], segmentation, feature extraction, and SVM classifier [ 10 , 11 ] have been implemented in the medical and veterinary fields.…”

The unmet potential of artificial intelligence in veterinary medicine

“…Pathogen detection from blood smears can be challenging when collected from exotic animals, particularly when commercial blood smear image analysis solutions are not available. One approach is to leverage CNNs: Kittichai et al 46 used a CNN to classify an avian malaria parasite from stained blood films with an accuracy of 97%.…”

A Deep Learning System for Recognizing and Recovering Contaminated Slider Serial Numbers in Hard Disk Manufacturing Processes

“…The excellent performance of the hybrid algorithms for developing an object detection and classification approach for classifying species and gender mosquito vectors [ 19 ] was previously published. Furthermore, an approach that combines an object detection model based on an object detection model with one of four classification models, namely Darknet, Darknet19, Darknet19-448, and Densenet-201, was effectively used to characterize the P. gallinaceum avian malaria blood stages [ 20 ]. In this work, the character detections for the first stage model were implemented on captured images using the object detection-based You-Only-Look-Once (YOLO) v4 CNN [ 21 ] (See Figure 3 ).…”