SomaDetect is helping transform how dairy farmers monitor and manage cow health and milk quality. This project focuses on developing advanced artificial intelligence (AI) tools to detect early signs of metabolic disorders like ketosis, and to optimize Energy-Corrected Milk (ECM) production across large and diverse dairy herds in Canada.
The agricultural challenge at the core of this work is the difficulty in consistently identifying high-performing cows and detecting early-stage health issues, particularly on large farms with multiple breeds and varied environmental conditions. Current monitoring methods often rely on delayed lab results and do not provide real-time, cow-level data and are insufficient to take proactive measures. When it comes to competing real-time milk analyzers, existing technologies are extremely capital intensive, limited to robotic farms, consume costly reagents and are expensive to maintain. Due to this, dairy farmers are searching for a suitable milk analyzer that is agnostic to their milking system and can provide the maximum return (both animal welfare and profitability) with the least amount of disruption.
SomaDetect’s proposed solution is a scalable, AI-powered system that leverages light scattering technology and machine learning to deliver cow-specific, real-time milk analysis during every milking. In this project, SomaDetect will be working with an industry leading Canadian dairy producer, Walker Dairy, to refine and validate existing machine learning models for two key metrics: the fat-to-protein ratio (FPR) and ECM in a commercial setting. It will also introduce a new early-warning system for subclinical ketosis—an often undetected metabolic disorder that negatively impacts milk yield and reproductive performance.
The core objective of this initiative is to advance SomaDetect’s technology from Technology Readiness Level (TRL) 6–7 to TRL 8 or greater, enabling full-system demonstration in an operational farm setting. To achieve this, the project team will conduct full-herd data collection, install new in-line sensors, and compare AI-generated shortlists against lab-verified results. They will co-develop on-farm protocols with farm personnel to incorporate these insights into daily operations. Results will be used to evaluate the return on investment, using on-farm financials and cow-level data to validate.
This project will directly benefit Canadian dairy producers by offering a precision livestock farming solution that is practical, cost-effective, and scalable. Dairy farmers will have an option that can provide meaningful nutrition and ketosis detection insights in real-time. It supports improved animal welfare, increased sustainability, and greater resilience within the dairy industry. By empowering producers with timely, actionable data, SomaElevate will help drive innovation and competitiveness in Canada’s agri-food sector, while contributing to national goals related to environmental stewardship, animal health, and technological advancement in agriculture. Phase 3 of our project directly quantifies the return on investment using actual results derived from the project on a large commercial dairy in Canada, which will serve as an example for operational (and profitability) improvements across the country.
For more information, please contact Nicholas Clermont, COO & Co-Founder, SomaDetect, at nicholasclermont@somadetect.com.
