The project offers a fundamentally new, integrative approach to the analysis of the composition of food products and dietary supplements, based on the synthesis of modern achievements in the field of artificial intelligence, biomedical informatics and regulatory control. Unlike existing solutions, the system is not limited to a superficial assessment of nutritional value, but implements a multi-level personalized interpretation of the composition, taking into account the individual parameters of the user and current regulations.
Текущее состояние
Currently, a scientific and algorithmic justification for personalized nutritional support has been developed, and a ready-to-implement digital tool with a high degree of practical and commercial readiness has been developed.
Проблема или Возможность
To date, none of the existing systems focused on the analysis of the composition of food products and dietary supplements has the range of functions necessary to implement a personalized, normatively substantiated and clinically relevant approach.
Решение (Продукт или Услуга)
The concept of an intelligent platform proposed within the project, combining personalized nutritional analytics, visual verification, normative-clinical validation and a trilingual interface, has practical significance for maintaining health and regulating the quality of consumed food products.
Конкуренты
Nutri-Score (European Union) is a nutritional value assessment system based on calories and the amount of individual macronutrients (fats, sugars, salt, fiber). However, Nutri-Score uses average models and does not take into account the health status of a particular consumer.
Yuka (France) is a mobile application that allows you to scan products and get composition assessments. It is based on a fixed rule base and does not interact with the user's medical or biomarker data.
Open Food Facts is an open international database of product composition that does not have personalized analytics and does not check labeling for compliance with legislation.
MyFitnessPal, Lifesum, Cronometer are calorie and nutrient trackers focused on self-monitoring and fitness goals, without regulatory or clinical validation.
Преимущества или дифференциаторы
For the first time in Kazakhstani practice, the project offers: intelligent interpretation of the composition of food products at an individual level; cross-validation of data according to the standards of various jurisdictions; the possibility of using the solution both in scientific research and in everyday life
Целевое назначение инвестиций
High-performance server hardware is critical for the successful development and stable support of an AI application. This requirement is due to the specifics of the operation of the key technological components of the system. In the process of developing and training artificial intelligence, resource-intensive tools such as Tesseract OCR for optical character recognition and the OpenCV library for image processing are used. These systems work with high-resolution photos (10+ megapixels), which places extremely high demands on computing power. To efficiently process such images, maximum-performance processors (CPUs) and specialized graphics processors (GPUs) optimized for parallel computing are required.
