Artificial intelligence applications in abiotic and biotic plant stress management: A comprehensive bibliometric and literature review

Abstract

Abiotic and biotic stress factors pose a significant threat to plant productivity and global food security. This review uses bibliometric analysis and literature synthesis to comprehensively examine the role of artificial intelligence (AI) and machine learning (ML) techniques in plant stress management. A total of 5369 publications retrieved from the Web of Science database between 2010 and 2024 were analysed using VOSviewer to evaluate publication trends, countries, institutions, and keywords. China, the US, and India were identified as the leading countries, with deep learning, convolutional neural networks, and image-based diagnostic methods emerging as key areas. The second phase revealed that DL architectures such as YOLO, EfficientNet, and Transformer, as well as methods like remote sensing and hyperspectral imaging, can accurately detect abiotic (drought, salinity, water, and heavy metals) and biotic (fungal, bacterial, and viral) stress. Machine learning (ML) algorithms such as support vector machine (SVM), random forest (RF), and artificial neural network (ANN) have also been found to be effective in stress prediction. However, challenges such as data imbalance, model interpretability, and high computational requirements persist. To address these issues, open-access datasets, low-cost and transparent models, multimodal sensing systems, and ethical AI approaches are recommended. This review contributes to the field by highlighting strategic and technical gaps in the development of scalable and sustainable AI-supported plant stress management systems.

Code and data availability

公開状態または取得可能な本文経路を確認できませんでした。

No evidence-backed public reproduction asset is currently recorded.