Adaptive neuro-fuzzy interface system and neural network modeling for the drying kinetics of instant controlled pressure drop treated parboiled rice

Ultrasound-assisted extraction of phytochemicals from green coconut shell: Optimization by integrated artificial neural network and particle swarm technique

This study employs artificial neural network (ANN) and particle swarm optimization (PSO) to maximize antioxidant and antimicrobial activity from green coconut shells. Phytochemical analysis was carried out on the extract obtained from ultrasound-assisted extraction performed at different combinations of time (10, 20, and 30 min), temperature (30, 35, and 40 °C ), and the ratio of solid-solvent (1:10, 1:20, and 1:30 g/ml). The presence of these bioactive compounds exhibits antimicrobial and antioxidant activities. Quantitative analysis showed that the total phenolic compounds ranged from 7.08 to 33.46 mg GAE/g, flavonoids ranged from 2.09 to 28.46 mg QE/g, tannins ranged from 70.5 to 141.09 mg TAE/g, and antioxidant activity of 49.98-66.1 %. The FTIR analysis detected the presence of C[bond, double bond]O, O-H, and C-H bonds. The optimized condition of ultrasound-assisted extraction (UAE) was compared with the optimized condition of the microwave. The result of ultrasound-assisted extraction was observed to be better than microwave-assisted extraction.

Effect of Euryale ferox seed shell extract addition on the in vitro starch digestibility and predicted glycemic index of wheat-based bread

The inhibitory effects of Euryale ferox seed shell extract (EFSSE) on the activity of α-amylase and α-glucosidase were studied. EFSSE (0.25 % to 2 %) was used to fortify bread and analyzed the in vitro starch digestibility (IVSD) digestion kinetics, and the predicted glycemic index (pGI) was estimated. The swarm intelligence supervised neural network (SISNN) technique was applied for the predictive simulation of digestion kinetics and pGI. Principal component analysis (PCA) with proportional odds modeling (POM) was used to find the most sensitive component based on the sensory attributes of bread. The inhibitory effect of EFSSE on α-amylase and α-glucosidase in terms of IC₅₀ was 62.95 and 52.06 μg/mL, respectively. Fortification of bread with EFSSE could affect loaf volume, hardness, and color. Euryale ferox seed shell extract could decreased the rate of hydrolysis of bread. EFSSE (2 %) had a strong inhibitory impact, as evidenced by the drop in glycemic index from 94.61 to 61.66. SISNN-based kinetics was much better as compared to mathematical modeling-based digestion kinetics. Findings of the present study have shown that EFSSE could be employed as an additive to produce lower glycemic index functional bread.

Application of Artificial Neural Networks, Support Vector, Adaptive Neuro-Fuzzy Inference Systems for the Moisture Ratio of Parboiled Hulls

Drying as an effective method for preservation of crop products is affected by various conditions and to obtain optimum drying conditions it is needed to be evaluated using modeling techniques. In this study, an adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN), and support vector regression (SVR) was used for modeling the infrared-hot air (IR-HA) drying kinetics of parboiled hull. The ANFIS, ANN, and SVR were fed with 3 inputs of drying time (0–80 min), drying temperature (40, 50, and 60 °C), and two levels of IR power (0.32 and 0.49 W/cm2) for the prediction of moisture ratio (MR). After applying different models, several performance prediction indices, i.e., correlation coefficient (R2), mean square error index (MSE), and mean absolute error (MAE) were examined to select the best prediction and evaluation model. The results disclosed that higher inlet air temperature and IR power reduced the drying time. MSE values for the ANN, ANFIS tests, and SVR training were 0.0059, 0.0036, and 0.0004, respectively. These results indicate the high-performance capacity of machine learning methods and artificial intelligence to predict the MR in the drying process. According to the results obtained from the comparison of the three models, the SVR method showed better performance than the ANN and ANFIS methods due to its higher R2 and lower MSE.