Declarations 362 • Code availability: Code of this study is available at https://github.com/363 fengchen025/PhenoAssistant/. 364 • Data availability: Data for demonstrating Case Study 1 can be requested from 365 http://phenotiki.com/. Data for training the computer vision model used in Case 366 Study 1 are publicly available at https://codalab.lisn.upsaclay.fr/competitions/367 8970. Data for demonstrating Case Study 2 are publicly available at https:// 368 zenodo.org/records/7938231. Data for demonstrating Case Study 3 are publicly 369 available at https://codalab.lisn.upsaclay.fr/competitions/13833.370 • Funding: This project was funded by the BBSRC grant BB/Y512333/1 371 “PhenomUK-RI: The
Open resource ↗pdf-raw-page:15 lines:1-84Unverified paper record
PhenoAssistant: A Conversational Multi-Agent AI System for Automated Plant Phenotyping
Research Square · 13 May 2025 · 10.21203/rs.3.rs-6430233/v1
Abstract
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 significantly lowering technical hurdles, PhenoAssistant underscores the promise of AI-driven methodologies to democratising AI adoption in plant biology.
Plant phenotyping relevance
植物フェノタイピングの画像解析ワークフローを自然言語で自動化するシステムを開発し、ケーススタディと評価タスクで検証しているため、方法が中心的である。
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's authors publicly release the PhenoAssistant analysis code (including chat logs for all case studies) on GitHub, and the paper uses public plant image datasets (CVPPP LSC for Case Study 1 model training; a public dataset for Case Study 3). Case Study 1 demonstration data is only available on request from the
iki.com/. Data for training the computer vision model used in Case 366 Study 1 are publicly available at https://codalab.lisn.upsaclay.fr/competitions/367 8970. Data for demonstrating Case Study 2 are publicly available at https:// 368 zenodo.org/records/7938231. Data for demonstrating Case Study 3 are publicly 369 available at https://codalab.lisn.upsaclay.fr/competitions/13833.370 • Funding: This project was funded by the BBSRC grant BB/Y512333/1 371 “PhenomUK-RI: The UK Plant and Crop Phenotyping Infrastructure”, and 372 Microsoft Accelerating Foundation Models Research (AFMR) grant: Agricultural 373 Foundation Models via Domain-Specific Pre-Training. 374 • Competing interests: The author
Open resource ↗pdf-raw-page:15 lines:1-84This is an automatically classified, unverified record. Curator approval is required before any resource enters the Catalog.