Research shows that the Smart24Food Atlas, a free digital tool developed for Argentina, successfully helps people identify foods 99.81% of the time and estimate portion sizes with an average error of only 1.34 grams. According to Gram Research analysis, when 108 young adults tested the atlas in a controlled study, they accurately matched food portions and retrieved correct food entries with minimal bias, suggesting the tool reliably improves dietary assessment accuracy for Argentine populations.
Researchers in Argentina created a free digital tool called Smart24Food Atlas that helps doctors and nutritionists understand what people eat by showing pictures of foods and recipes with their nutritional information. The tool includes nearly 1,000 different foods photographed in realistic portion sizes, making it easier for people to remember and describe their meals accurately. When 108 young adults tested the tool by identifying foods and matching portion sizes in a controlled study, they successfully found the right foods 99.8% of the time and correctly estimated portion sizes with minimal error. This resource could help doctors give better nutrition advice and researchers study eating patterns more accurately across Argentina.
Key Statistics
A 2026 controlled study of 108 participants found that the Smart24Food Atlas achieved 99.81% success in food retrieval tasks, with participants correctly identifying foods from a digital atlas containing 998 distinct foods and 3,273 photographs.
The Smart24Food Atlas demonstrated high accuracy in portion-size estimation, with participants’ weight estimates averaging only 1.34 grams off from actual portions across 1,404 visual matching observations, showing no systematic bias in either direction.
The Smart24Food Atlas contains nutrient-composition information for up to 53 different nutritional components across 998 foods and standardized recipes commonly consumed in Argentina, making it a comprehensive resource for dietary assessment.
The Quick Take
- What they studied: Whether a new digital picture atlas of Argentine foods and recipes could help people accurately identify what they eat and estimate how much they consumed
- Who participated: 108 participants, mostly young adults at a university, who completed tasks identifying packaged foods and matching portion sizes to photographs
- Key finding: The atlas successfully helped participants identify foods 99.81% of the time and estimate portion sizes with an average error of only 1.34 grams, showing it works reliably for dietary assessment
- What it means for you: If you’re working with a doctor or nutritionist in Argentina, this tool could make it much easier to accurately track what you eat and help your healthcare provider give you better nutrition advice based on precise information
The Research Details
Researchers created a free digital atlas containing photographs of 998 different foods and recipes commonly eaten in Argentina, along with detailed nutritional information for each item. The atlas includes 3,273 different food photographs showing realistic portion sizes and presentations. To test whether the tool actually works, they recruited 108 university students and asked them to complete 20 tasks: seven tasks involved identifying packaged products from pictures, and 13 tasks involved looking at a photograph of a food portion and estimating how much it weighed. The researchers watched participants complete these tasks in a controlled laboratory setting to measure accuracy.
The study used advanced statistical methods to analyze the results. For identifying foods correctly, researchers used a type of statistical analysis that accounts for the fact that some foods might be harder to identify than others, while some participants might be better at the task. For portion-size estimation, they measured how close participants’ estimates were to the actual weight and looked for any patterns in their errors, such as consistently overestimating or underestimating.
This approach allowed researchers to test the atlas under ideal conditions before recommending it for real-world use in doctor’s offices and nutrition research studies.
Understanding what people actually eat is one of the biggest challenges in nutrition science and healthcare. Without accurate information about diet, doctors cannot give good nutrition advice, and researchers cannot study how food affects health. The biggest problem is that people often forget what they ate or cannot accurately describe portion sizes. By providing realistic photographs of foods in standard portions, this atlas removes guesswork and helps people remember and communicate their eating habits more accurately. Testing the tool in a controlled setting first ensures it works before doctors and researchers start using it with real patients.
This study has several strengths: it tested a large number of foods (998 different items), used a reasonable sample size (108 participants), and measured performance objectively by comparing participant estimates to actual weights. The researchers also used appropriate statistical methods that account for variation between foods and between people. However, the study tested only young adults in a university setting, which may not represent all Argentines. The controlled laboratory conditions are ideal for testing but may not reflect how well the tool works in real-world situations when people are busy or distracted. The researchers acknowledge these limitations and call for future studies with more diverse age groups and in everyday settings.
What the Results Show
The Smart24Food Atlas performed exceptionally well in this controlled test. When participants tried to find and identify foods from the atlas, they succeeded 99.81% of the time, meaning they found the correct food in nearly every attempt. This extremely high success rate suggests the atlas is well-organized and the food photographs are clear and recognizable.
For the portion-size matching tasks, which tested whether people could accurately estimate how much food they were looking at, results were also very good. On average, participants’ estimates were only 1.34 grams off from the actual weight. To put this in perspective, that is less than the weight of a single grain of rice. Importantly, there was no evidence that participants systematically overestimated or underestimated portions. Some people guessed slightly high, others slightly low, but overall the errors balanced out with no consistent bias.
The researchers also found that the accuracy of portion-size matching did not depend on how much food was actually served. Whether participants were estimating a small portion or a large portion, their accuracy remained consistent. This is important because it means the tool works reliably across different serving sizes.
The study found that some foods were easier to identify and match than others, which is expected. Different types of foods showed different levels of difficulty, but the atlas performed well across all categories tested. The researchers noted that the specific way a food was presented in the photograph (such as whether it was on a plate or in a bowl) did not significantly affect whether people could match portions accurately. This suggests the atlas is flexible enough to work in various real-world eating situations.
According to Gram Research analysis, previous studies of food atlases in other countries have reported similar or slightly lower accuracy rates for portion-size estimation. This Argentine atlas appears to perform as well as or better than existing tools used in other countries. The extremely high food-retrieval rate (99.81%) is particularly impressive and suggests the atlas design and organization are effective. However, most previous studies, like this one, tested tools in controlled laboratory settings rather than in everyday life, so direct comparisons are somewhat limited.
The biggest limitation is that all participants were young adults at a university, so results may not apply to older people, children, or people with different education levels. The study took place in a controlled laboratory where people had time to carefully complete tasks, which is very different from real life when people are eating quickly or distracted. The study did not test whether the atlas works well for people with vision problems or those unfamiliar with technology. Additionally, the atlas was tested with people looking at photographs on a screen, but in real use, people might be using it on smartphones or tablets, which could affect performance. The researchers acknowledge these limitations and recommend future studies in more diverse groups and in real-world eating situations.
The Bottom Line
Healthcare providers and nutrition researchers in Argentina should consider using the Smart24Food Atlas as a tool to help patients and study participants accurately report what they eat. The high accuracy demonstrated in this study suggests it will improve the quality of dietary information collected. Confidence level: High for controlled settings like doctor’s offices or research studies. Confidence level: Moderate for everyday use, pending further testing in real-world conditions.
This tool is most useful for people working with doctors or nutritionists in Argentina who need to track their diet for health reasons, such as managing diabetes, heart disease, or obesity. It is also valuable for nutrition researchers studying eating patterns in Argentine populations. People who speak Spanish and are familiar with Argentine foods will benefit most. The tool may be less useful for people unfamiliar with technology or those with significant vision problems, though these questions need further study.
The atlas can improve dietary assessment immediately when used in clinical or research settings. However, benefits to health depend on what doctors or nutritionists do with the dietary information collected. If the improved accuracy leads to better nutrition advice and people follow that advice, health improvements could take weeks to months to become noticeable, depending on the health goal.
Frequently Asked Questions
How accurate is the Smart24Food Atlas for estimating portion sizes?
The atlas achieved an average error of only 1.34 grams in portion-size estimation during testing with 108 participants. Estimates ranged from 33.44 grams underestimated to 30.76 grams overestimated, with no consistent bias toward over or underestimation.
Can I use the Smart24Food Atlas to track my daily nutrition intake?
Yes, the atlas contains nutritional information for up to 53 components per food item, allowing you to track calories, protein, fiber, and other nutrients. However, accuracy depends on correctly identifying foods and estimating portions using the provided photographs.
Is the Smart24Food Atlas available for free?
Yes, the Smart24Food Atlas is an open-access resource, meaning it is free to use. It was designed specifically for foods and recipes commonly eaten in Argentina and includes 998 different foods with realistic portion-size photographs.
Who should use the Smart24Food Atlas?
Healthcare providers, nutritionists, and researchers in Argentina can use the atlas to help patients and study participants accurately report their diet. It is most useful for people managing health conditions through diet or participating in nutrition research studies.
Does the Smart24Food Atlas work for all age groups?
The atlas was tested only with young adults at a university, so its effectiveness for children, older adults, or other populations is unknown. Further research is needed to confirm it works equally well across all age groups in Argentina.
Want to Apply This Research?
- Users can photograph their meals and compare them to the Smart24Food Atlas photographs to log portion sizes and calories. Track daily totals for three key nutrients: protein, fiber, and added sugars. Record this information weekly to identify eating patterns.
- When eating a meal, open the atlas and find the closest matching photograph before eating. Estimate your portion size using the atlas as a reference. Log the food and portion size in the app immediately while the meal is fresh in your memory. This creates a real-time dietary record rather than trying to remember meals at the end of the day.
- Review your logged meals weekly to identify which foods and portions are most common in your diet. Compare your actual portions to recommended serving sizes shown in the atlas. Track whether your portion sizes change over time and whether your nutrient intake meets your health goals. Share reports with your doctor or nutritionist quarterly to adjust nutrition recommendations based on accurate data.
This research describes the development and testing of a digital food atlas tool under controlled laboratory conditions with young adults. Results may not apply to all populations, ages, or real-world eating situations. The Smart24Food Atlas is a research and clinical tool designed to improve dietary assessment accuracy, not a substitute for professional medical or nutritional advice. Always consult with a qualified healthcare provider or registered dietitian before making significant changes to your diet or using dietary assessment tools for medical purposes. The accuracy of any dietary assessment tool depends on proper use and honest reporting by the user.
This research translation is published by Gram Research, the science division of Gram, an AI-powered nutrition tracking app.