Validation of Digital Slide Scanning and a Convolutional Neural Network for the Detection of Intestinal Parasites in Human Stool Samples

A recent publication by Dr. Alexander Oberli and the University of Bern provides a prospective validation of digital microscopy and AI for routine intestinal parasite detection 👇

The study evaluated the Grundium Ocus 40 scanner in tandem with our partners' solution, Techcyte Human Fecal Wet Mount (HFW) algorithm.

In non-endemic zones, where the majority of stool samples are negative, manual screening is often redundant and uncomfortable.

This AI-assisted workflow optimizes laboratory throughput by allowing technicians to concentrate on suspicious slides while reliably excluding negative cases.

𝗞𝗲𝘆 𝗣𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲 𝗦𝘁𝗮𝘁𝗶𝘀𝘁𝗶𝗰𝘀:
⇒ 97.6% slide-level agreement with expert light microscopy after classifier optimization.
⇒ 98.1% agreement in prospective routine diagnostics (κ = 0.915).
⇒ High intra- and inter-run reproducibility across diagnostic cycles.
⇒ Reduced manual workload via AI-based pre-classification while maintaining high diagnostic accuracy.

It's been a privilege to work with Dr. Oberli and Techcyte's team.

Access the full study here: https://lnkd.in/eggfJs-d