In a tropical village, a lab technician peers through a microscope eyepiece, hunting for tiny worm eggs in a drop of fluid. The work is monotonous, the eyes tire, and it's easy to miss a parasite. Yet the outcome determines whether a person gets treatment.

A new approach hands the search to a neural network that understands both text and images. Trained on images of parasite eggs, it now circles each egg with a frame, like a vigilant assistant that never blinks. In tests, the Florence model found eggs more accurately than previous object recognition methods.

Such a system could become part of a simple device: a drop on a slide, a snapshot, and an answer within a minute. In places without parasitology specialists, this could prevent chronic diseases and stunted growth in children.