Treetracker: intelligent plant production optimization software with machine learning integration

Overview

TreeTracker is a cloud-capable software platform designed to optimize micropropagation-based plant production through the integration of supervised machine learning capabilities. Developed for high-value horticultural crop producers, TreeTracker addresses a critical gap in the market: existing micropropagation software is costly and lacks predictive functionality. By embedding ML within a purpose-built production environment, TreeTracker enables growers to derive actionable insights from operational data, peer-reviewed research, and institutional knowledge, supporting more efficient, data-driven decisions across the entire micropropagation workflow.

The platform combines a Design of Experiments module with a Research Knowledge Database (RKB), creating a structured environment where ML algorithms can be trained and applied to real-world production scenarios. The result is a tool that helps laboratories and commercial growers improve plant growth outcomes, increase production output, and reduce costs, particularly for crops such as grapes, berries, hops, and orchard species.

Technical specifications

Key features:

• Supervised machine learning integration for predicting optimal growth conditions, media formulations, and acclimatization outcomes
• Design of Experiments module that streamlines experimental planning and reduces the resources required for trial-and-error testing
• Research Knowledge Database (RKB) consolidating institutional knowledge including 482 media recipes, 219 experiments, and 114 laboratory standard operating procedures, alongside variables drawn from peer-reviewed literature
• Cloud-capable architecture enabling flexible deployment and access across production sites
• Context-aware recommendations that output the most efficient practices based on user-defined production contexts
• Compatibility with existing micropropagation operations, allowing live data acquisition to continuously train and refine ML models

The ML capabilities draw on demonstrated approaches from prior research, including the use of artificial neural networks to simulate in vitro plant growth and predict optimal conditions for growth and acclimatization. The platform is designed to model effects of plant hormones and media recipes, supporting optimization across multiple crop types.

Technology readiness level

TreeTracker is positioned at an advanced development stage, with the core cloud-capable software platform already in place. The proposed validation pathway begins with standardization of micropropagation process parameters and population of the Research Knowledge Database. Following this, ML algorithms will be trained using the RKB combined with live data from operational micropropagation activities. Finally, the ML capabilities will be evaluated at each stage of the micropropagation process, with direct comparisons to traditional micropropagation practices to confirm efficacy. This staged validation approach supports a clear path toward commercial deployment in horticultural production environments.