Application and Prospects of Artificial Intelligence in Schistosomiasis Prevention and Control
TL;DR
Future directions encompass multimodal diagnostic integration, transfer learning with foundation models, large language model (LLM)-assisted decision-making decision-making, and a "human-animal-environment" intelligent prevention and control system.
Abstract
Schistosomiasis is a serious zoonotic parasitic disease affecting approximately 250 million people worldwide. Despite significant progress in China, challenges remain in snail surveillance, the low sensitivity of traditional diagnostics, and imprecise resource allocation. Recent advances in artificial intelligence offer new solutions. In snail monitoring, CNNs achieve over 90% accuracy in image recognition, and deep learning combined with remote sensing accurately identifies snail habitats. For diagnosis, mobile microscopy integrated with deep learning enables automated on-site egg detection. In strategy optimization, machine learning models (e.g., XGBoost) quantitatively evaluate intervention cost-effectiveness, and decision support systems simulate integrated control effects. Challenges include data quality, accessibility, and ethical privacy concerns. Future directions encompass multimodal diagnostic integration, transfer learning with foundation models, large language model (LLM)-assisted decision-making decision-making, and a "human-animal-environment" intelligent prevention and control system.