ARTIFICIAL INTELLIGENCE-GENERATED DIETS IN COMPARISON: A NUTRIENT PROFILING APPROACHES OF POPULAR DIET MODELS

Öz

Objective: AI systems could significantly enhance dietetics, particularly in creating personalized nutrition plans. The ability to analyze large datasets can increase the effectiveness of these systems in formulating nutritious and healthy dietary patterns. Nevertheless, it remains unclear whether these systems can be effectively employed in the formulation of nutritionally balanced and healthy diets. This study aimed to evaluate the effectiveness of vegan and ketogenic diets generated using two different AI tools (ChatGPT and COPILOT). Methods: Each AI tool was tasked with the design of 24 meals for both the vegan and ketogenic diets, totaling 96 meals. Subsequently, a comparative analysis was conducted to evaluate the nutritional characteristics and health-promoting aspects of the diets generated using these tools. The statistical analyses were conducted using the SPSS 24.0 software. Results: Significant differences were identified in carbohydrate, vitamin B6, calcium, and magnesium levels between ketogenic diets generated by AI tools. Among the vegan diets, only the percentage of carbohydrates and fats exhibited statistically significant differences. Moreover, the NRF 15.3 index, atherogenic index, SFA/USFA ratio, ω-6/ω-3 ratio, and DIAAS demonstrated significant differences between the diets generated by the AI tools. No significant differences in these parameters were observed between the vegan diets. Conclusion: Although AI tools have demonstrated potential advantages in dietetics, further research, including diet plans formulated by dietitians, is necessary to confirm the effectiveness of these findings.

Anahtar Kelimeler

• Artificial intelligence
• ChatGPT
• COPILOT
• Vegan diet
• Ketogenic diet
• NRF 15.3

Yapay Zeka Destekli Diyetlerin Karşılaştırılması: Popüler Diyetlere Besin Ögesi Profilleme Yaklaşımı

Öz

Amaç: Yapay zeka sistemleri, bireye özgü beslenme planları ile diyetetik alanında anlamlı rol oynama potansiyeline sahiptir. Büyük veri setlerini analiz etme kabiliyeti, bu sistemlerin besleyici ve sağlıklı diyet örüntülerini formüle etmedeki etkinliğini artırabilir. Ancak, bu sistemlerin besleyici ve sağlıklı diyetlerin geliştirilmesinde etkili bir şekilde kullanılıp kullanılamayacağı henüz belirsizdir. Bu bağlamda çalışma, iki farklı yapay zeka aracı (ChatGPT ve COPILOT) tarafından oluşturulan vegan ve ketojenik diyetlerin etkinliğini değerlendirmeyi amaçlamıştır. Yöntemler: Bu çalışmada, yapay zeka araçlarının her birine hem vegan hem de ketojenik diyetler için 24 adet öğün tasarımı görevi verilerek toplamda 96 öğün elde edilmiştir. Bu araçlar tarafından oluşturulan diyetlerin besin ögeleri, besleyici özellikleri ve sağlığı geliştirici yönleri karşılaştırmalı bir analiz ile değerlendirilmiştir. İstatistiksel analizler SPSS 24.0 programı kullanılarak gerçekleştirilmiştir. Bulgular: Yapay zeka araçları tarafından oluşturulan ketojenik diyetler arasında karbonhidrat, B6 vitamini, kalsiyum ve magnezyum miktarlarında önemli farklılıklar gözlenmiştir. Vegan diyetler arasında ise sadece karbonhidrat ve yağ yüzdeleri istatistiksel olarak anlamlı düzeyde farklılık göstermiştir. Ayrıca, yapay zeka araçları tarafından oluşturulan diyetler arasında Nutrient Rich Foods (NRF) 15.3 indeksi, aterojenik indeks, doymuş/doymamış yağ asidi oranı (SFA/USFA), omega-6/omega-3 oranı ve protein kalite skoru (DIAAS) önemli farklılıklar göstermiştir. Vegan diyetler arasında ise, bu parametrelerde anlamlı farklılıklar gözlenmemiştir. Sonuç: Yapay zeka araçları diyetetik alanında potansiyel avantajlar göstermiş olsa da, bulguların etkinliğini doğrulamak için diyetisyenler tarafından formüle edilen diyet planlarının da dahil edildiği daha kapsamlı araştırmalara ihtiyaç vardır.

Anahtar Kelimeler

• Yapay zeka
• ChatGPT
• COPILOT
• Vegan diyet
• Ketojenik diyet
• NRF 15.3

Kaynakça

1. [1] Athanasian, C.E., Lazarevic, B., Kriegel, E.R., Milanaik, R.L. (2021). Alternative diets among adolescents: facts or fads? Current Opinion in Pediatrics, 33(2), 252-9.
2. [2] Bogataj Jontez, N., Kenig, S., Šik Novak, K., Petelin, A., Jenko Pražnikar, Z., Mohorko, N. (2023). Habitual low carbohydrate high fat diet compared with omnivorous, vegan, and vegetarian diets. Frontiers in Nutrition, 10, 1106153.
3. [3] Borkent, J.W., Grootswagers, P., Linschooten, J., Roodenburg, A.J.C., Ocké, M., de van der Schueren, M.A.E. (2024). A vegan dietary pattern is associated with high prevalence of inadequate protein intake in older adults; a simulation study. The Journal of Nutrition, Health and Aging, 28(10), 100361.
4. [4] Brand-Miller, J., Buyken, A.E. (2020). The relationship between glycemic index and health. Nutrients, 12(2), 536.
5. [5] Brouns, F. (2018). Overweight and diabetes prevention: is a low-carbohydrate-high-fat diet recommendable? European Journal of Nutrition, 57(4), 1301-12.
6. [6] Christodoulides, S.S,, Neal, E.G., Fitzsimmons, G., Chaffe, H.M., Jeanes, Y.M., Aitkenhead, H. et al. (2012). The effect of the classical and medium chain triglyceride ketogenic diet on vitamin and mineral levels. Journal of Human Nutrition and Dietetics, 25(1), 16-26.
7. [7] Darmon, N., Drewnowski, A. (2015). Contribution of food prices and diet cost to socioeconomic disparities in diet quality and health: a systematic review and analysis. Nutrition Reviews, 73(10), 643-60.
8. [8] Elorinne. A.L., Alfthan, G., Erlund, I., Kivimäki, H., Paju, A., Salminen, I. et al. (2016). Food and nutrient intake and nutritional status of finnish vegans and non-vegetarians. PLOS One, 11(2), e0148235.

9. [9] Food and Agriculture Organization of the United Nations Report of an FAO Expert Consultation. (2013). Dietary protein quality evaluation in human nutrition. Erişim adresi: http://www.fao.org/ag/ humannutrition/35978-02317b979a686a57aa4593304ffc17f06.pdf. [Erişim tarihi: 14 Ekim 2024].
10. [10] Freire, R. (2020). Scientific evidence of diets for weight loss: Different macronutrient composition, intermittent fasting, and popular diets. Nutrition, 69, 110549.
11. [11] Fuehrlein, B.S., Rutenberg, M.S., Silver, J.N., Warren, M.W., Theriaque, D.W., Duncan, G.E. et al. (2004). Differential metabolic effects of saturated versus polyunsaturated fats in ketogenic diets. The Journal of Clinical Endocrinology & Metabolism, 89(4), 1641-5.
12. [12] Han, J.H., Lee, H.J., Cho, M.R., Chang, N., Kim, Y., Oh, S.Y. et al. (2014). Total antioxidant capacity of the Korean diet. Nutrition Research and Practice, 8(2), 183-91.
13. [13] Hieronimus, B., Hammann, S., Podszun, M.C. (2024). Can the AI tools ChatGPT and Bard generate energy, macro- and micro-nutrient sufficient meal plans for different dietary patterns? Nutrition Research, 128, 105-14.
14. [14] Jalali, S., Heidari, Z., de Courten, B., Rashidkhani, B. (2023). Dietary total antioxidant capacity and odds of breast cancer: a case-control study. Nutrition and Cancer, 75(1), 302-9.
15. [15] Jeong, H.Y., Moon, Y.S., Cho, K.K. (2024). ω-6 and ω-3 Polyunsaturated Fatty Acids: Inflammation, Obesity and Foods of Animal Resources. Food Science of Animal Resources, 44(5), 988-1010.
16. [16] Kaiser, J., van Daalen, K.R., Thayyil, A., Cocco, M., Caputo, D., Oliver-Williams, C.A. (2021). Systematic Review of the Association Between Vegan Diets and Risk of Cardiovascular Disease. The Journal of Nutrition, 151(6), 1539-52.
17. [17] Kenig, S., Petelin, A., Poklar Vatovec, T., Mohorko, N., Jenko-Pražnikar, Z. (2019). Assessment of micronutrients in a 12-wk ketogenic diet in obese adults. Nutrition, 67, 110522.
18. [18] Keys, A., Anderson J, Grande F. (1957). Prediction of serum-cholesterol responses of man to changes in fats in the diet. The Lancet, 270(7003), 959-66.
19. [19] Keys, A, Anderson, J.T., Grande, F. (1965). Serum cholesterol response to changes in the diet: II. The effect of cholesterol in the diet. Metabolism - Clinical and Experimental, 14(7), 759-65.
20. [20] Keys, A., Anderson, J.T., Grande, F. (1965). Serum cholesterol response to changes in the diet: IV. Particular saturated fatty acids in the diet. Metabolism - Clinical and Experimental, 14(7), 776-87.
21. [21] Lara-Arevalo, J., Laar, A., Chaparro, M.P., Drewnowski, A. (2024). Nutrient-dense African indigenous vegetables and grains in the FAO food composition table for Western Africa (WAFCT) identified using nutrient-rich food (NRF) scores. Nutrients, 16(17), 2985.
22. [22] Lee, Y.M., Kim, S.A., Lee, I.K., Kim, J.G., Park, K.G., Jeong, J.Y. et al. (2016). Effect of a brown rice based vegan diet and conventional diabetic diet on glycemic control of patients with type 2 diabetes: A 12-week randomized clinical trial. PLoS One, 11(6), e0155918.
23. [23] Liu, L., Hu, Q., Wu, H., Xue, Y., Cai, L., Fang, M. et al. (2016). Protective role of n6/n3 PUFA supplementation with varying DHA/EPA ratios against atherosclerosis in mice. The Journal of Nutritional Biochemistry, 32, 171-80.
24. [24] Ludwig, D.S. (2002). The Glycemic IndexPhysiological Mechanisms Relating to Obesity, Diabetes, and Cardiovascular Disease. Journal of the American Medical Association, 287(18), 2414-23.
25. [25] Mansoor N, Vinknes KJ, Veierød MB, Retterstøl K. (2016). Effects of low-carbohydrate diets v. low-fat diets on body weight and cardiovascular risk factors: a meta-analysis of randomised controlled trials. British Journal of Nutrition, 115(3), 466-79.
26. [26] Mathai, J.K., Liu, Y., Stein, H.H. (2017). Values for digestible indispensable amino acid scores (DIAAS) for some dairy and plant proteins may better describe protein quality than values calculated using the concept for protein digestibility-corrected amino acid scores (PDCAAS). British Journal of Nutrition, 117(4), 490-9.
27. [27] Paknahad, Z., Moosavian, S.P., Jervekani, Z.T., Hasanzadeh, A., Hashemi, M. (2021). Dietary total antioxidant capacity and severity of stenosis in patients with coronary artery disease. International Journal for Vitamin and Nutrition Research, 91(3-4), 235–41.
28. [28] Papastratis, I., Stergioulas, A., Konstantinidis, D., Daras, P., Dimitropoulos, K. (2024). Can ChatGPT provide appropriate meal plans for NCD patients? Nutrition, 121, 112291.
29. [29] Perna, M., Hewlings, S. (2022). Saturated fatty acid chain length and risk of cardiovascular disease: A systematic review. Nutrients, 15(1), 30.
30. [30] Saunders, A.V., Davis, B.C., Garg, M.L. (2013) Omega-3 polyunsaturated fatty acids and vegetarian diets. Medical Journal of Australia, 199, 22-26.
31. [31] Türkiye Beslenme Rehberi (TÜBER) (2022). Sağlık Bakanlığı, Halk Sağlığı Genel Müdürlüğü, Sağlık Bakanlığı Yayın No:1031, Ankara.
32. [32] USDA. United States Department of Agriculture. (2018). USDA Food Composition Databases. https://fdc.nal.usda.gov. Erişim tarihi: 22 Ekim 2024.
33. [33] Ventriglio, A., Sancassiani, F., Contu, M.P., Latorre, M., Di Salvatore, M., Fornaro, M. et al. (2020). Mediterranean diet and its benefits on health and mental health: A literature review. Clinical Practice and Epidemiology in Mental Health, 16(1), 156-64.
34. [34] Ülker, İ., Ayyıldız, F. (2021). Artificial intelligence applications in nutrition and dietetics. Intelligent Systems with Applications, 4(2), 125-7.
35. [35] Łuszczki, E., Boakye, F., Zielińska, M., Dereń, K., Bartosiewicz, A., Oleksy, Ł. et al. (2023). Vegan diet: Nutritional components, implementation, and effects on adults' health. Frontiers in Nutrition, 10, 1294497.