HEALTHGUARD analyses different physiological parameters from a user to identify hidden features and thus make personalized nutritional recommendations.
Value Proposition: There is little knowledge on how the data provided by smartwatches and other similar wearables can be transformed in actionable information with a direct positive effect on health. HealthGuard, a machine learning software, is specially intended to cover this unmet need by:
- analyzing the data provided by the instrument.
- understanding the physiological meaning of these data to detect early signs of deviation from healthy ranges.
- providing nutritional interventions (e.g. foods, supplements, ingredients) that can amend such deviations.
What’s new: HealthGuard is a machine learning framework designed and developed by Topazium using the most advanced AI technologies to promote healthy nutritional habits tailored to their unique moment in life.
Key insights: The physiological parameters analysed by HealthGuard to build its customized recommendations may come from different sources. Usually, such data is collected by smartwatches or similar gadgets. However, HealthGuard can also analyse data from specialized medical sources like blood tests or devices like smart beds in hospitals or instruments continuously collecting data from patients.
How it works: Using machine learning analysis, HealthGuard can detect deviations of physiological parameters from healthy trends, even if their values are still in normal ranges. Once the deviation is detected, this information is used to propose customized nutritional interventions to amend the deviation. Moreover, since the information is being continuously received, the effect of these interventions on the observed deviations can be judged and modified accordingly, giving rise to a continuous improvement process. HealthGuard utilizes cutting-edge neural networks to connect food or nutritional supplement composition with physiological effects. This enables HealthGuard to select foods and ingredients that will cause a physiological effect purposed to amend the deviation that had been detected. Finally, making use of generative AI models, the tool combines the selected ingredients in creative, personalized recipes designed to elicit the intended effect in the user.
Why it matters: The overall population is increasingly aware of the impact that salutary habits and invigorating routines have on health preservation. The use of digital devices such as smartwatches and other similar wearables are contributing to such awareness, but there is little knowledge on how the data provided by these devices can be transformed in actionable information with a direct positive effect on health. Based on self-learning tools and generative AI models, HealthGuard provides the user with creative recipes utilizing foods or nutritional supplements whose components will have a personalized positive effect on their health. These recommendations are based on the data delivered by the personal device in real time, therefore adapted to the specific needs of the users according to their actual circumstances.
HOW TO RUN THE EXPERIMENT
Access: available through a dedicated app and/or via an API.
Input data: Connect your smartwatch, smartphone, or other wearable device to our HealthGuard platform.
Running your experiment: Expected run time for each image: less than 5 minutes
Output: You will receive a PDF by email with a list of recommended foods, their ingredients, their metapharmacological properties, and examples of recipes you can make with the recommended foods.
AWARDS
- Best Health Data Driven-Innovation – Finalist. European digital mindset awards 2023, DES#2023.
- Artificial Intelligence and Advanced Technologies in Health 2023– Winner. Madrid City Hall.