AI-Supported Microscopy Improves Detection of Intestinal Parasite Infections

Soil-transmitted helminths (STHs), such as roundworm, whipworm, and hookworm, are among the most common neglected tropical diseases, affecting over 600 million people globally. These infections are particularly harmful to children in low-resource settings, where they contribute to malnutrition, anemia, and hinder physical and mental development. The difficulty in diagnosing STHs accurately in such settings has been a major obstacle to effective treatment. Traditional manual microscopy often fails to detect light infections, leading to misdiagnosis or delayed treatment. Now, a new diagnostic solution that combines artificial intelligence (AI) with portable digital microscopy can improve the detection of these infections.

This solution was created by researchers at Karolinska Institutet (Stockholm, Sweden) and international collaborators after analyzing 704 stool samples using three methods: manual microscopy, fully autonomous AI, and expert-verified AI. The AI-based systems were inspired by the need for faster and more accurate diagnostics in areas with limited resources. The expert-verified AI approach, where local experts confirm AI findings in under a minute, was found to be the most accurate. The AI system is capable of analyzing a sample in about 15 minutes, with expert confirmation taking just one minute, making it an efficient tool for use in primary healthcare settings.

Image: AI microscopy can improve parasite detection within primary health care (Photo courtesy of Joar von Bahr)

The AI solution was tested and validated by comparing its performance to manual microscopy in detecting hookworm, whipworm, and roundworm infections. The expert-verified AI system achieved detection rates of 92% for hookworm, 94% for whipworm, and 100% for roundworm, significantly outperforming manual microscopy. These findings were published in Scientific Reports, showing that AI combined with portable microscopy can provide accurate and scalable diagnostics at the point of care. This is particularly crucial as the prevalence of STH declines, necessitating more sensitive methods for disease monitoring. The researchers plan to further refine this technology for wider use in global health efforts.

“This research shows the potential of combining portable imaging with AI to overcome long-standing diagnostic challenges in global health,” said Johan Lundin, professor at the Department of Global Public Health, Karolinska Institutet.

Related Links:
Karolinska Institutet