Related Experiment Video
Updated: Aug 13, 2026
09:24
Use of In Vivo Imaging to Screen for Morphogenesis Phenotypes in Candida albicans Mutant Strains During Active Infection in a Mammalian Host
Published on: October 12, 2022
Fungal morphotype detection and quantification in microscopic images with TU_MyCo-vision: a user-friendly deep
Kartik J Deopujari1, Matthias Schmal1, Caroline Danner1
1Institute of Chemical, Environmental and Bioscience Engineering, TU Wien, Gumpendorfer Strasse 1a, Wien, 1060, Austria.
Fungal Biology and Biotechnology
|August 12, 2026
Summary
Researchers developed TU_MyCo-Vision, a YOLO-based tool for identifying 13 fungal morphotypes in microscopic images. This automated approach overcomes limitations of manual observation for studying fungal morphological diversity.
Area of Science:
- Mycology
- Computational Biology
- Image Analysis
Background:
- Fungal morphological switching is crucial for adaptation but challenging to quantify.
- Manual microscopic analysis is labor-intensive, prone to bias, and difficult to scale.
- Existing image-based tools lack broad applicability across diverse fungal taxa.
Purpose of the Study:
- To develop an automated, versatile tool for identifying fungal morphotypes.
- To enable high-throughput analysis of fungal morphological diversity.
- To provide a standardized method for fungal phenotype quantification.
Main Methods:
- Developed TU_MyCo-Vision using Ultralytics YOLO (You Only Look Once) object detection.
- Trained a YOLOv11m model on 1,504 custom annotated images of Aureobasidium pullulans.
- Integrated a graphical user interface for data analysis and visualization.
Main Results:
- The best model (Zulu_s3) achieved 73.4% precision and 66.5% recall across 13 morphotypes.
- The tool successfully analyzed fungal morphotype distribution and co-occurrence.
- Validated across different fungal genera (Candida albicans, Komagataella phaffi, Aspergillus niger).
Conclusions:
- TU_MyCo-Vision offers a scalable and objective method for fungal morphotype identification.
- The tool demonstrates potential for comparative studies of fungal morphological diversity across taxa.
- Automated image analysis can significantly advance fungal research.
