Knowledge graph-based digital phenotyping platform for accelerating crop resistance development
A knowledge graph-based digital phenotyping platform that integrates multispectral imaging, 3D models, and omics data to accelerate crop resistance development. It applies graph algorithms and machine learning to relate phenotype signals to resistance factors, enabling root-cause exploration and transfer of insights across species. The platform supports flexible dataset expansion and compatibility with the DELTA imaging platform for non-destructive time-series imaging.
Overview
This solution applies knowledge graph technology to digital crop phenotyping, enabling faster and more flexible identification of plant resistance traits. The approach builds on experience creating digital plant twins and AI-based analysis tools for crop trials, translating that expertise into a graph-based framework that integrates multispectral imaging, 3D models, spectral vegetation indices, morphological characteristics, environmental data, and omics data into a unified digital phenotype.
The approach enables researchers to explore root causes of resistance development and transfer insights between species, reducing service costs. By combining graph modelling with existing AI techniques, the platform captures complex spatial and temporal relationships between data points, providing a more complete picture of plant dynamics than conventional methods.
Technical specifications
Core capabilities:
- Integration of multispectral images, 3D models, spectral vegetation indices, and morphological features into a unified graph database
- Application of graph algorithms and machine learning techniques including community detection, similarity, node embedding, node classification and regression, and link prediction
- Correlation of phenotype data with omics and environmental datasets to identify resistance factors
- Flexible platform architecture that allows datasets to be easily expanded and integrated
- Compatibility with the DELTA imaging platform for non-destructive time-series imaging
Performance benchmarks from existing methods:
- Spectral vegetation indices demonstrate statistically significant differences between susceptible and resistant lines up to 50% faster than manual scoring by trained agronomists
- AI models have shown up to 80% improvement in time to discovery for identifying performing lines
Technology readiness level
The technology is at TRL 4-6, with validation planned over a one-year period. The study will begin with a precursor analysis using existing trial data. Following this, crops including oilseed rape and wheat will be grown and infected on-site, imaged over time using the DELTA platform, and combined with omics and environmental data. Graph-based models will then be benchmarked against classical and deep learning approaches to evaluate enhancements or losses in execution.
About FOTENIX
Fotenix is an agri-tech company that provides AI-driven crop monitoring and analytical solutions designed to improve farm profitability and sustainability. The company’s core technology utilizes three-dimensional multispectral imaging and machine learning to analyze crop health, enabling early detection of pest and disease outbreaks, nutrient deficiencies, and plant stress. By integrating this non-destructive imaging hardware with a cloud-based platform, Fotenix allows growers to transition from reactive management to precise, data-driven interventions. The system is designed to be easily installed and integrated into existing agricultural workflows, supporting applications such as automated quality control, crop scouting, and smart resource targeting for machinery like sprayers and harvesters.
These tools address critical challenges in modern agriculture, including labor shortages for manual monitoring, high operational costs, and the need to reduce chemical inputs. By providing actionable insights that improve resource efficiency and quality, Fotenix helps farmers increase productivity while minimizing environmental impact. Founded in 2018 as a University of Manchester spinout, the company has partnered with various agricultural organizations and research centers to scale its operations. It has secured support through government grants, venture capital, and industry-led programs to advance its product development and expand its presence in the horticulture and plant breeding markets.
delivered to your inbox.