Nutritional enhancement and physiological responses to papaya (Carica papaya L.), turmeric (Curcuma longa L.), moringa (Moringa oleifera Lam.) - enriched biscuits and coconut (Cocos nucifera L.) in Wistar rats: artificial neural network modeling

Widespread malnutrition and the demand for nutrient-rich diets have prompted growing interest in enriching foods with plant-derived additives. However, how these complex diet formulations interrelate and influence physiological outcomes remains poorly understood. This research aimed to use an artificial neural network (ANN) to evaluate the efficacy of plant derivatives ( Carica papaya L.; Curcuma longa; Moringa oleifera; and Cocos nucifera ) in improving physiological and nutritional responses. The Wistar rats ( Rattus norvegicus ) were divided into 16 groups. The negative control (NC) received baseline feed; the positive control (PC) received synthetic multivitamin at 2 mL/kg/day. Treatments 1–14 were administered biscuits enriched with various concentrations of plant derivatives. For treatments 5, 11, 13, and 14, drinking fluid was replaced with coconut water. The rats were observed over a 56-day period, followed by collection of blood specimens for analysis of hematological parameters, lipid profiles, and hepatic and renal function markers. The feed intake, weight gain, and feed conversion ratio (FCR) were evaluated. All the tests were conducted in accordance with approved standards. The ANN utilized 20 features, trained employing the Levenberg–Marquardt optimization algorithm, and the data were split into 70% training, 15% validation, and 15% testing sets. The findings showed that plant derivatives significantly improved the levels of nutritional and bioactive compounds in enriched biscuit samples ( p ≤ 0.05). Similarly, the developed ANN models displayed outstanding predictive accuracy, with excellent regression coefficients for all outputs ( R 2 ≥ 0.900). The results revealed that the experimental groups treated with multi-additive treatments showed substantially better physiological Responses. The activities of superoxide dismutase, catalase, and glutathione peroxidase were elevated by 125.1%, 66.0%, and 117.9%, respectively, whereas malondialdehyde levels decreased by 70.5%. The body weight gain increased from 113.33 g in the NC to 188.33 g in T14, whereas the FCR declined from 9.96 to 7.19. Furthermore, substantial decreases were observed across all key serum biomarkers: alanine aminotransferase (42.85–21.04 U/L), aspartate aminotransferase (107.17–69.00 U/L), alkaline phosphatase (162.00 to 119.17 U/L), and creatinine (0.99 to 0.63 mg/dL). These results highlight the potential of biscuits enriched with natural additives as functional foods enriched in nutrients and bioactive compounds, and indicate that ANN may serve as a complementary approach for modeling complex interactions between diet and physiological outcomes.