Automated diagnosis of schistosomiasis by using faster R-CNN for egg detection in microscopy images prepared by the Kato–Katz technique

One of the biggest concerns in the area of public health is caused by human intestinal parasites, which are found largely in tropical countries. The diagnosis of these parasitic diseases is done through physiological symptoms and fecal examination. Often, few professionals are available and able to perform this type of examination, which is considered time-consuming, requires trained personnel, prone to errors, and can cause eye strain in the specialist. In this paper, we investigate the use of the faster R-CNN object detection method to identify eggs of Schistosoma mansoni , forming a system to aid decision making in the diagnosis of fecal examination. A real database was built with 66 images prepared by the Kato–Katz method. Online and offline data augmentation techniques were used to obtain a larger number of samples. As a result, the proposed solution reached an average precision value of 0.765 for an IoU (intersection over union) of 0.50. The results and applicability of the system are promising and may be used in public health programs to assist health professionals in the diagnosis and monitoring of schistosomiasis in endemic areas.