This review offers a comprehensive analysis of the expanding role of artificial intelligence (AI) in mycological research. It highlights how AI supports the identification of fungal taxa at both microscopic and macroscopic levels, facilitates the detection of toxic species, contributes to ecological mapping, aids in discovering new fungal bioproducts, and enhances precision agriculture practices. Advanced techniques such as convolutional neural networks, you only look once, U-shaped network (UNet), and residual network have demonstrated strong performance in classification, segmentation, and object detection tasks. Additionally, explainable AI (XAI) methods like gradient-weighted class activation mapping (Grad-CAM), local interpretable model-agnostic explanations (LIME), and SHapley additive exPlanations (SHAP) improve the transparency of model decisions by providing visual and quantitative insights, thereby fostering greater clarity in scientific applications. Practical tools such as MUSH-AI showcase how these technologies can be integrated to improve mushroom cultivation through predictive models, optimisation of environmental conditions, and early detection of diseases. For high-value fungi like truffles, AI has proven valuable in identifying suitable habitats using satellite imagery, analysing mycorrhizal relationships, and interpreting spectral data, offering more efficient alternatives to traditional techniques such as searches with trained dogs. However, the wider application of AI faces several barriers, including a lack of robust training datasets, variability in annotation quality, limited generalisability of models, and insufficient transparency during field deployment. Overcoming these challenges will require the development of standardised, ecologically rich fungal image databases, strong institutional partnerships, and significant investment in infrastructure. In conclusion, AI is becoming a core element in modern fungal research, streamlining species identification, mycotoxin analysis, resource discovery, and conservation efforts through effective and scalable tools. With ongoing advancements in data quality and interpretability, AI is poised to shape the future of integrated and sustainable mycology.
Ekinci et al. (Tue,) studied this question.