The code for this research, as well as the chat logs and generated outputs of the case studies and evaluations, are available at Github [ https://github.com/vios-s/PhenoAssistant/ ] 78 .
Open resource ↗vios-s/PhenoAssistant · lines:224-268Unverified paper record
A conversational multi-agent AI system for automated plant phenotyping.
Nature Communications · 3 Apr 2026 · 10.1038/s41467-026-71090-y
Abstract
Plant phenotyping increasingly relies on (semi-)automated image-based analysis workflows to improve its accuracy and scalability. However, many existing solutions remain overly complex, difficult to reimplement and maintain, and pose high barriers for users without substantial computational expertise. To address these challenges, we introduce PhenoAssistant: a pioneering AI-driven system that streamlines plant phenotyping via intuitive natural language interaction. PhenoAssistant leverages a large language model to orchestrate a curated toolkit supporting tasks including automated phenotype extraction, data visualisation and automated model training. We validate PhenoAssistant through several representative case studies and a set of evaluation tasks. By lowering technical hurdles, PhenoAssistant underscores the promise of AI-driven methodologies to democratising AI adoption in plant biology.
Plant phenotyping relevance
植物表現型抽出を自然言語で自動化するAIシステムの開発であり、ツールとワークフローが研究の中心です。代表的ケーススタディと評価タスクによる検証も行っています。
abstractwe introduce PhenoAssistant: a pioneering AI-driven system that streamlines plant phenotyping via intuitive natural language interaction.
abstractPhenoAssistant leverages a large language model to orchestrate a curated toolkit supporting tasks including automated phenotype extraction, data visualisation and automated model training.
abstractWe validate PhenoAssistant through several representative case studies and a set of evaluation tasks.
Code and data availability
The paper deposits its PhenoAssistant analysis code (with chat logs and generated outputs) on GitHub, and uses public phenotyping datasets: the CVPPP2017 leaf segmentation challenge data (case study 1 training/evaluation) and the CVPPA@ICCV'23 WW2020 winter wheat nutrient-deficiency dataset (case study 3), both on Coda
The data used for training and evaluating the computer vision model used in case study 1 are publicly available from the CVPPP2017 Leaf Segmentation Challenge dataset (A1 and A4 subsets) at CodaLab [ https://codalab.lisn.upsaclay.fr/competitions/8970 ].
Open resource ↗CodaLab · CVPPP2017 · lines:224-268The winter wheat data used in case study 3 are publicly available from the CVPPA@ICCV'23: image classification of nutrient deficiencies in winter wheat and winter rye dataset (WW2020 subset) at CodaLab [ https://codalab.lisn.upsaclay.fr/competitions/13833 ].
Open resource ↗CodaLab · WW2020 · lines:224-268This is an automatically classified, unverified record. Curator approval is required before any resource enters the Catalog.