A transparent explanation of the methodology, data sources, AI approach and limitations behind each tool. We believe you deserve to understand how results are generated.
When you submit a food photo, it is sent to an AI language model with vision capabilities. The AI analyses the image and attempts to identify the food items present based on visual characteristics including colour, texture, shape and typical presentation. The AI draws on its training to recognise common foods across many cuisines.
Once the AI identifies the food, it cross-references that identification against USDA FoodData Central — a database maintained by the United States Department of Agriculture containing nutritional data for a large range of foods. The AI generates an estimate based on a typical serving or portion of the identified food.
The word "estimate" is important and used deliberately. The scanner cannot measure your food — it analyses a two-dimensional photograph. Several factors make precise measurement impossible from a photo alone:
When you enter an ingredient name or E-code, the AI cross-references it against reference information from dietary authorities and food additive registries. The tool screens against five dietary categories: Halal, Vegan, Vegetarian, Gluten-Free and Dairy-Free.
Results are shown as one of three categories:
When you describe symptoms, the AI generates general health information relevant to those symptoms and organises it by a broad urgency level — Low, Medium or Critical. This is designed to help you understand whether something may need prompt medical attention.
Urgency levels are generated by the AI based on general clinical triage principles. They are broad indicators only and should not be treated as a definitive assessment of your individual situation. Many symptoms can have multiple causes with varying degrees of seriousness.
BMI is calculated using a standard formula: weight in kilograms divided by height in metres squared (kg/m²). For imperial measurements, the tool converts to metric before calculating. This is the same formula used by the WHO, NHS and CDC.
BMI is a population-level screening tool and has well-documented limitations when applied to individuals. It does not account for muscle mass, bone density, age, sex, or ethnicity. A highly muscular person may have a high BMI without excess body fat. Older adults may have a healthy BMI but still carry excess body fat. For these reasons, BMI should be used as one indicator among several, not as a standalone measure of health.
Try the BMI Calculator →You provide your daily calorie target, dietary preference (Halal, Vegan, Gluten-Free, Dairy-Free or balanced), health goal (lose weight, maintain or gain muscle) and the plan duration. The AI generates a suggested meal plan with breakfast, lunch and dinner for each day, including estimated calories and protein per meal.
The tool screens named foods or ingredients against the 14 major allergens regulated under EU Food Information Regulation 1169/2011 and FDA major food allergen requirements. The AI identifies allergens that are typically present in the named food based on standard ingredients.
Afiya AI uses an AI language model to generate results. The AI draws on reference information from publicly available sources that were part of its training, as well as the following specifically referenced databases and guidelines:
USDA FoodData Central — The US Department of Agriculture's food composition database. Used as a reference for nutritional estimates in the calorie scanner. Published at fdc.nal.usda.gov.
JAKIM (Department of Islamic Development Malaysia), IFANCA (Islamic Food and Nutrition Council of America), HMC (Halal Monitoring Committee UK), EU E-number registry — Used as reference guidelines for halal ingredient screening. Note: Afiya AI is not affiliated with any of these organisations and does not issue certification on their behalf.
EU Food Information Regulation 1169/2011 — 14 major allergens. US FDA — Major food allergen requirements. Used as reference for allergen screening.
WHO ICD-10-CM — International Classification of Diseases, used as a general reference framework for organising health information. WHO BMI classifications — Used for BMI category definitions.
Vegan Society ingredient guidance and GFFS (Gluten-Free Food Service) standards used as reference for vegan and gluten-free ingredient screening.