Hyperspectral tissue analysis for in-season plant health diagnostics
Expansion of Persistence Data Mining's hyperspectral diagnostic platform from soil and post-harvest grain analysis into in-season plant tissue testing. The solution delivers rapid, non-destructive nutrient and health assessments for corn, soybean, wheat, alfalfa, canola, and hay, enabling year-round use of hyperspectral hardware and a complete suite of soil, food, and plant health diagnostics for farmers, agronomists, and industry partners.
Overview
Persistence Data Mining has built a hyperspectral imaging platform, marketed as Soilytics, that converts spectral data into actionable soil nutrient and grain quality measurements without traditional wet-lab chemistry. This project extends those proprietary algorithms into plant tissue analysis during the growing season, adding root ball and leaf diagnostics to the existing soil health and post-harvest grain offerings. The combined suite gives farmers, agronomists, and industry partners a single rapid, reliable testing workflow that can be deployed across most of the calendar year, maximizing hardware utilization and lowering the cost per sample for end users.
Technical specifications
Core technology:
- Contact-probe hyperspectral sensor covering the 350–2500 nm wavelength range for capturing spectral signatures from homogenized samples
- Machine-learning models trained on more than 400,000 field-correlated samples to convert spectral readings into quantifiable nutrient values
- Algorithms that account for moisture, geological characteristics, soil texture, and color to improve predictive accuracy (RMSEP)
- Planned extension of the algorithm library to interpret tissue spectral data for in-season plant health diagnostics
Target crops for tissue validation:
- Corn, soybean, wheat, alfalfa, canola, and hay
- Minimum of 200 chemical-lab-correlated tissue samples per crop
- Integration of soil and tissue datasets to build a unified soil and plant health database
Delivery model: Non-destructive, near real-time analysis intended to replace or complement traditional chemical tissue testing for prescription mapping and crop health decisions.
Technology readiness level
The soil nutrient and post-harvest grain analysis capabilities are commercially validated, supported by extensive field trials and prior industry recognition including an R&D 100 Award. The tissue analysis extension is at an early-to-mid development stage, with prior external work suggesting feasibility but requiring new algorithm development, sample collection, and chemical correlation across six crops before commercial readiness. The current project focuses on building the correlated dataset and validating predictive accuracy for plant tissue diagnostics.
About Persistence Data Mining
Persistence Data Mining Inc., which rebranded as Soilytics Inc. in November 2024, is an AgTech company that provides advanced soil fertility diagnostics. The company developed the Soilytics platform, which utilizes proprietary hyperspectral imaging technology and advanced algorithms to replace traditional, time-intensive chemical soil testing with faster, physics-based analysis. This technology is designed to create accurate, granular soil nutrient maps, allowing farmers, agronomists, and land managers to optimize fertilizer application, improve crop yields, and enhance soil health while reducing environmental impacts such as runoff.
By offering near real-time insights into soil health and nutrient management, the company helps agricultural stakeholders navigate challenges such as soil degradation and rising input costs. Their solution is intended to streamline the process of creating prescription maps, moving beyond the inefficiencies of manual lab-based sampling. The company, which has been recognized with industry honors such as an R&D 100 Award, works with partners in the precision agriculture market to integrate its digital diagnostic capabilities into broader farming operations.
delivered to your inbox.