Comparative accuracy of smartphone apps and a generative AI tool for carbohydrate counting: An independent bicentric study
Authors' conclusion
In this controlled hospital-based evaluation, GluciCheck, DiabHealth, and EkiYou provided the most accurate and consistent carbohydrate estimations. ChatGPT-5, despite not being designed for nutritional analysis, achieved intermediate accuracy comparable to several dedicated tools, whereas Gluroo showed the greatest variability. These findings highlight the potential of both structured manual systems and emerging AI-based technologies to support carbohydrate counting in T1D management.
Study context
- Population
- 246 hospital meals estimated by trained medical students acting as mock patients; reference values by expert dietitians from weighed food items and standardized photographs
- Participants
- 246 hospital meals
- Comparison
- Four carbohydrate-counting smartphone apps vs ChatGPT-5
- Outcome
- Absolute error in carbohydrate estimation
Methodological detailsbicentric, standardized hospital-based environment
- Bicentric
- Standardized hospital-based environment
Performance
As reported in the abstract
P < 0.05
Secondary outcomes
| Outcome | Values |
|---|---|
| Meals estimated within ±20 g | GluciCheck 77% · EkiYou 67% · DiabHealth 80% · Gluroo 50% · ChatGPT-5 73% |
| Spearman correlation with reference | GluciCheck ρ 0.78 · EkiYou ρ 0.71 |
- Spearman correlation with reference — Moderate correlations for DiabHealth, Gluroo and ChatGPT-5
- ρ ≈ 0.50
Bland-Altman agreement
- Narrowest limits of agreement for GluciCheck and EkiYou; widest for Gluroo
- DiabHealth and ChatGPT-5 showed minimal bias and no proportional error
Original publication
Comparative accuracy of smartphone apps and a generative AI tool for carbohydrate counting: An independent bicentric study. Diabetes Obes Metab. Published 22 December 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →