Deep-learning assisted detection and quantification of (oo)cysts of Giardia and Cryptosporidium on smartphone microscopy images

The consumption of microbial-contaminated food and water is responsible for the deaths of millions of people annually. Smartphone-based microscopy systems are portable, low-cost, and more accessible alternatives for the detection of Giardia and Cryptosporidium than traditional brightfield microscopes. However, the images from smartphone microscopes are noisier and require manual cyst identification by trained technicians, usually unavailable in resource-limited settings. Automatic detection of (oo)cysts using deep-learning-based object detection could offer a solution for the limitation. Automatic detection of (oo)cysts using deep-learning-based object detection could offer a solution for the limitation. We evaluate the performance of three state-of-the-art object detectors to detect (oo)cysts of Giardia and Cryptosporidium} on a custom dataset that includes both smartphone and brightfield microscopic images from the vegetable samples. Faster RCNN, RetinaNet, and you only look once (YOLOv8s) deep-learning models were employed to explore their efficacy and limitations. Our results show that while the deep-learning models perform better with the brightfield microscopy image dataset than the smartphone microscopy image dataset, the smartphone microscopy predictions are still comparable to the prediction performance of non-experts.