Book a demo
GET STARTED
Case Study

AI-powered nutrition monitoring for elite athletes at CAR Sant Cugat

The Centre d’Alt Rendiment de Sant Cugat supports elite athletes who require precise, continuous and personalized nutritional monitoring as part of their training and performance programs.
Learn more

Background & Objectives

The Centre d’Alt Rendiment de Sant Cugat supports elite athletes who require precise, continuous and personalized nutritional monitoring as part of their training and performance programs.

Traditional food-intake monitoring was manual, slow and subjective, limiting the ability of nutrition professionals to obtain structured, real-time information on what athletes actually consume.

The objective of this first phase was to evaluate LogMeal’s AI and computer-vision technology to:

  • Automate food recognition from images
  • Extract nutritional information from real meal records
  • Support nutritionists with objective intake data
  • Validate the technical feasibility of the system in a high-performance sports environment
  • Prepare future phases for scalable athlete nutrition monitoring

Highlighted data

  • 150+ number of athletes/professionals involved
  • All sports disciplines included

 

 

Initial assessment & digital monitoring with Logmeal

Initial Assessment

  • Automate food recognition from images
  • Existing monitoring depended on manual food records and professional interpretation
  • Nutritionists needed faster and more objective dietary information
  • Meal-image collection was used to assess recognition quality and operational feasibility
  • The project focused on Phase 1: technical validation of LogMeal algorithms in the CAR environment

Digital Monitoring with LogMeal

  • Athletes or staff capture images of meals
  • LogMeal AI identifies dishes, ingredients and food groups
  • The system generates nutritional information, including calories, macronutrients and micronutrients
  • Meal data supports structured follow-up by nutrition professionals
  • The validation establishes the basis for future integration with CAR systems through API-based workflows

Highlighted data

  • 81,000+ number of images collected / month
  • 98%+ recognition accuracy on tray accuracy
  • 90%+ whether LogMeal App, API, Platform or Kiosk was used

 

Professional workflow & future implementation

Nutrition Professional Workflow

  • Review athlete meal records using objective visual data
  • Compare actual intake with nutritional targets
  • Identify dietary gaps, deviations and improvement opportunities
  • Support personalized recommendations based on real consumption
  • Reduce dependence on memory-based or self-reported food logs

Roadmap for Scalable Implementation

  • Automatic association between meal trays and individual users
  • Platform visualization for nutrition teams
  • Personalized recommendations for athletes
  • Quantity detection and portion-size estimation
  • Integration with clinical history or athlete records
  • API integration with CAR systems and potential Catalan healthcare-system workflows

Highlighted data

  • Integration with clinical history or athlete records
  • Continue acquiring meal images from the spring–summer menu to strengthen dish recognition
  • Import the CAR athlete database into the LogMeal system
  • Enable voluntary athlete identification using the CAR card via barcode or NFC
  • Automatically associate each scanned tray with the corresponding athlete profile
  • Integrate CAR recipes and nutritional data to provide real-time intake analysis
  • Give the nutrition team structured intake data without manual forms or mobile use during meals.

Results and value of Logmeal in the program

Outcomes

  • Improved patient engagement and adherence through an easy-to-use, visual food logging system.
  • Technical validation of AI-based food recognition in an elite sports environment
  • A more objective basis for athlete dietary monitoring
  • Potential reduction in manual work for nutrition professionals
  • Better structured nutritional data for follow-up and decision-making
  • A foundation for scalable and personalized nutrition monitoring at CAR

LogMeal Technology Value in the Program

  • AI-powered food recognition from images
  • Automated nutritional analysis covering macro- and micronutrients
  • Professional oversight through structured data workflows
  • API-first architecture for integration with existing systems
  • An adaptable roadmap for high-performance sports environments

Conclusion

The CAR Sant Cugat project demonstrates how LogMeal can transform meal images into structured and actionable nutritional data for elite sports professionals.

The first phase validates the technical basis for a more objective, scalable and personalized model of athlete nutrition monitoring, supporting both professional decision-making and future integration with CAR’s digital ecosystem.

partners and collaborators