Synthetic Fungi Datasets: A Time-Aligned Approach
A. Rani, D. O. Arroyo, P. Durdevic
TL;DR
This work addresses the need for time-resolved fungal imaging data by introducing a time-aligned synthetic dataset that simulates fungal growth from spores to hyphae and ultimately to dense mycelium networks. A structured generation model combines spatial, temporal, and probabilistic rules, including a transition parameter $T$ and counts $S(t)$, $H(t)$, and $M(t)$, to produce consistent growth across frames. The dataset offers a scalable, DL-ready resource with recursive branching, Poisson-distributed sub-branching, and environment-aware variations suitable for classification, segmentation, and growth-prediction tasks. By enabling automated fungal analysis and domain adaptation between synthetic and real microscopy data, the work supports advancements in agriculture, medicine, and industrial mycology through AI-driven methodologies.
Abstract
Fungi undergo dynamic morphological transformations throughout their lifecycle, forming intricate networks as they transition from spores to mature mycelium structures. To support the study of these time-dependent processes, we present a synthetic, time-aligned image dataset that models key stages of fungal growth. This dataset systematically captures phenomena such as spore size reduction, branching dynamics, and the emergence of complex mycelium networks. The controlled generation process ensures temporal consistency, scalability, and structural alignment, addressing the limitations of real-world fungal datasets. Optimized for deep learning (DL) applications, this dataset facilitates the development of models for classifying growth stages, predicting fungal development, and analyzing morphological patterns over time. With applications spanning agriculture, medicine, and industrial mycology, this resource provides a robust foundation for automating fungal analysis, enhancing disease monitoring, and advancing fungal biology research through artificial intelligence.