Automated Diagnosis of Newcastle Disease in Chickens Using Fecal Images Based on Machine Learning

Newcastle disease is a highly contagious viral disease that affects poultry, and it is causing significant economic losses in the poultry industry worldwide. Early diagnosis of the disease is critical for effective disease control and management. By analyzing feces images, farmers can identify the types of infectious diseases that attack chickens and take necessary actions to increase production yields. Machine learning classifiers have shown promise in automated disease diagnosis, including Newcastle disease in chickens using fecal images. In this paper, we present a machine learning (ML)-based classification for automated diagnosis of Newcastle disease in chickens using fecal images. The proposed approach includes data augmentation to address the issue of limited data, followed by the application of various statistical features to the images for feature extraction. The output of the feature extraction step is then applied to different machine learning classifiers, including random forest (RF), decision tree (DT), support vector machine (SVM), K-nearest neighbor (KNN), and logistic regression (LR). Experimental results demonstrate that the random forest classifier outperforms the other algorithms, achieving an accuracy of 96.12%, a precision of 94.01%, a sensitivity of 99.19, and an F-measure of 96.53% in the automated diagnosis of Newcastle disease in chickens using fecal images.