Artificial neural network modelling of the antioxidant activity and phenolic compounds of bananas submitted to different drying treatments

Effects of Different Heating Treatments on the Antioxidant Activity and Phenolic Compounds of Ecuadorian Red Dacca Banana

The banana is a tropical fruit characterized by its composition of healthy and nutritional compounds. This fruit is part of traditional Ecuadorian gastronomy, being consumed in a wide variety of ways. In this context, unripe Red Dacca banana samples and those submitted to different traditional Ecuadorian heating treatments (boiling, roasting, and baking) were evaluated to profile their phenolic content by ultra-high-performance liquid chromatography coupled to high-resolution mass spectrometry (UHPLC-HRMS) and the antioxidant activity by ORAC, ABTS, and DPPH assays. A total of sixty-eight phenolic compounds were identified or tentatively identified in raw banana and treated samples, highlighting the content in flavonoids (flavan-3-ols with 88.33% and flavonols with 3.24%) followed by the hydroxybenzoic acid family (5.44%) in raw banana samples. The total phenolic compound content significantly decreased for all the elaborations evaluated, specifically from 442.12 mg/100 g DW in fresh bananas to 338.60 mg/100 g DW in boiled (23.41%), 243.63 mg/100 g DW in roasted (44.90%), and 109.85 mg/100 g DW in baked samples (75.15%). Flavan-3-ols and flavonols were the phenolic groups most affected by the heating treatments, while flavanones and hydroxybenzoic acids showed higher stability against the heating treatments, especially the boiled and roasted samples. In general, the decrease in phenolic compounds corresponded with a decline in antioxidant activity, evaluated by different methods, especially in baked samples. The results obtained from PCA studies confirmed that the impact of heating on the composition of some phenolic compounds was different depending on the technique used. In general, the heating processes applied to the banana samples induced phytochemical modifications. Even so, they remain an important source of bioactive compounds for consumers.

Modeling of Drying Kinetics of Banana (Musa spp., Musaceae) Slices with the Method of Image Processing and Artificial Neural Networks

In this study, modeling of thin banana slices dried on 316 stainless steel shelves is carried out in an oven working with serial controlled and concentric blower-resistor couple. Changes occurred in banana slices (area and color) during drying process have been recorded by a camera. Additionally, weight has been measured with a load cell which is under the shelf and energy consumption has been measured with electricity consumption meter which is tied to energy input. The main aim of the study is to conduct the drying process of banana slices according to the data obtained from camera. Besides, obtained data have been tested with a powerful modeling technique like Artificial Neural Networks (ANN), and it has been seen that drying process could be modeled according to the data obtained from camera. Energy consumption data have been added in order to increase the performance of ANN and strengthen the modeling. Thus, an automatic drying system that can learn by itself using only a camera without any other sensors will be installed. This has been caused an increase in performance. However, it is obvious that it increases cost. According to the results of modeling process, 99% of “goodness of fit” has been obtained by using the change in banana slices and the number of pixels. It has been found that the developed model performed adequately in predicting the changes of the moisture content. Thus, it has been available to control the food drying process with a digital camera.

A Comprehensive Review of Mathematical Modeling for Drying Processes of Fruits and Vegetables

Drying fruits and vegetables is a procedure of food preservation with simultaneous heat, mass, and momentum transfer, which increases the shelf life of the food product. The aim of this review was to provide an overview of the researches on mathematical modeling for drying of fruits and vegetables with the special emphasis on the computational approach. Various heat-mass transport models, their applications, and modern drying technologies to the food industry have been reported in this study. Computational fluid dynamics, a new approach for solving heat and mass transfer problems, increases the accuracy of the predicted values. To investigate the parameters of drying needs a significant amount of time as well as costly laboratory and experimental efforts. Therefore, computational modeling could be an effective alternative to experimental approaches. This review will be beneficial for future studies in drying processes, especially for modeling, analysis, design, and optimization of food science and food engineering.

Artificial Neural Networks for Predicting Food Antiradical Potential

Using an artificial neural network (ANN), the values of the antiradical potential of 1315 items of food and agricultural raw materials were calculated. We used an ANN with the structure of a “multilayer perceptron” (MLP) and with the hyberbolic tangent (Tanh) as an activation function. Values reported in the United States Food and Nutrient Database for Dietary Studies (FNDDS) were taken as input to the analysis. When training the ANN, 60 parameters were used, such as the content of plastic substances, food calories, the amount of mineral components, vitamins, the composition of fatty acids and additional substances presented in this database. The analysis revealed correlations, namely, a direct relationship between the value of the antiradical potential (ARP) of food and the concentration of dietary fiber (r = 0.539) and a negative correlation between the value of ARP and the total calorie content of food (r = −0.432) at a significance level of p < 0.001 for both values. The average ARP value for 10 product groups within the 95% CI (confidence interval) was ≈23–28 equivalents (in terms of ascorbic acid) per 1 g of dry matter. The study also evaluated the range of average values of the daily recommended intake of food components (according to Food and Agriculture Organization—FAO, World Health Organization—WHO, Russia and the USA), which within the 95% CI, amounted to 23.41–28.98 equivalents per 1 g of dry weight. Based on the results of the study, it was found that the predicted ARP values depend not only on the type of raw materials and the method of their processing, but also on a number of other environmental and technological factors that make it difficult to obtain accurate values.

Drying kinetics of two fruits Portuguese cultivars (Bravo de Esmolfe apple and Madeira banana): An experimental study

Air convective dehydration was carried out at a laboratory scale using two fruits of cultivars produced in different regions of Portugal: Bravo de Esmolfe apple, from Beiras province, and Cavendish banana, from Madeira Island. Fresh fruits were dried in a tray drier with a hot airstream at different temperatures (35, 40, 45, and 50 °C) and velocity of 1.6 m s⁻¹. Drying rate curves were obtained using a simple mathematical approach applied to the moisture content curves adjusting linear and polynomial functions. Different drying rate stages were noticed in the experiments made with apples (one constant drying rate period followed by two falling drying rate periods), while in the case of the banana the constant drying rate period was not perceived, being dried entirely during a unique falling drying rate period. As expected, the constant drying rate value obtained at the beginning of the experiments with apples is higher when these were conducted at higher temperatures, changing from 8.103 to 14.474 g m⁻² s⁻¹ when the airstream temperature increases from 35 to 50 °C. The correspondent critical moisture contents in the Bravo de Esmolfe apples, at the instant the constant drying rate period stops and the drying rate starts to fall, slightly decreases from 4.800 to 4.134 kg_water/kg_{dry solid}.

This study explored for the first time the drying behavior of these two important fruits that have been increasingly used in the food industry in Portugal, giving important information for the industrialization of its production.

Kluai Hin (Musa sapientum Linn.) peel as a source of functional polyphenols identified by HPLC-ESI-QTOF-MS and its potential antidiabetic function

To date, information on the polyphenolic composition of Kluai Hin banana peel and pulp and the potential antidiabetic activity of its major active compounds is limited. This study aimed to identify polyphenols in extracts of fresh and freeze-dried Kluai Hin banana peel and pulp (methanol:water; M:W, 80:20 for flavonoids and acetone:water:acetic acid; A:W:A, 50:49:1 for phenolic acids) by RP-HPLC-DAD and HPLC-ESI-QTOF-MS. Additionally, inhibition of α-amylase and α-glucosidase activities was investigated with crude extracts from Kluai Hin banana peel and pulp, and compared with its major polyphenols ((+)-catechin, (-)-epicatechin and gallic acid) and the antidiabetic drug acarbose. (-)-Gallocatechin was the most abundant polyphenol and was detected in all fresh and freeze-dried pulp and peel extracts by RP-HPLC-DAD. Furthermore, unidentified polyphenol peaks of Kluai Hin were further explored by HPLC-ESI-QTOF-MS. The A:W:A fresh peel extract contained more total phenolic content (811.56 mg GAE/100 g) than the freeze-dried peel (565.03 mg GAE/100 g). A:W:A extraction of the fresh and freeze-dried peel of exhibited IC₅₀ values for α-amylase activity 2.66 ± 0.07 mg/ml and 2.97 ± 0.00 mg/ml, respectively, but its inhibitory activity was lower than acarbose (IC₅₀ = 0.25 ± 0.01 mg/ml). Peel extracts inhibited α-glucosidase activity, whereas pulp extracts had no effect. In addition, all standards, except gallocatechin, activated α-amylase activity, while, gallocatechin inhibited α-glucosidase activity better than acarbose. Therefore, we propose a further investigation into the use of Kluai Hin banana peel as a potential functional food for the management of postprandial glycaemic response to reduce diabetes risk and in the management of diabetes with a commercial drug.

Rapid quantification of dissolved solids and bioactives in dried root vegetable extracts using near infrared spectroscopy

Artificial neural networks (ANN) were developed for prediction of total dissolved solids, polyphenol content and antioxidant capacity of root vegetables (celery, fennel, carrot, yellow carrot, purple carrot and parsley) extracts prepared from the (i) fresh vegetables, (ii) vegetables dried conventionally at 50 °C and 70 °C, and (iii) the lyophilised vegetables. Two types of solvents were used: organic solvents (acetone mixtures and methanol mixtures) and water. Near-infrared (NIR) spectra were recorded for all samples. Principal Component Analysis (PCA) of the pre-treated spectra using Savitzky-Golay smoothing showed specific grouping of samples in two clusters (1st: extracts prepared using methanol mixtures and water as the solvents; 2nd: extracts prepared using acetone mixtures as the solvents) for all four types of extracts. Furthermore, obtained results showed that the developed ANN models can reliably be used for prediction of total dissolved solids, polyphenol content and antioxidant capacity of dried root vegetable extracts in relation to the recorded NIR spectra.

Extraction of Phenolic Compounds with Antioxidant Activity from Strawberries: Modelling with Artificial Neural Networks (ANNs)

This research study focuses on the evaluation of the total phenolic compounds (TPC) and antioxidant activity (AOA) of strawberries according to different experimental extraction conditions by applying the Artificial Neural Networks (ANNs) technique. The experimental data were applied to train ANNs using feed- and cascade-forward backpropagation models with Levenberg-Marquardt (LM) and Bayesian Regulation (BR) algorithms. Three independent variables (solvent concentration, volume/mass ratio and extraction time) were used as ANN inputs, whereas the three variables of total phenolic compounds, DPPH and ABTS antioxidant activities were considered as ANN outputs. The results demonstrate that the best cascade- and feed-forward backpropagation topologies of ANNs for the prediction of total phenolic compounds and DPPH and ABTS antioxidant activity factors were the 3-9-1, 3-4-4-1 and 3-13-10-1 structures, with the training algorithms of trainlm, trainbr, trainlm and threshold functions of tansig-purelin, tansig-tansig-tansig and purelin-tansig-tansig, respectively. The best R² values for the predication of total phenolic compounds and DPPH and ABTS antioxidant activity factors were 0.9806 (MSE = 0.0047), 0.9651 (MSE = 0.0035) and 0.9756 (MSE = 0.00286), respectively. According to the comparison of ANNs, the results showed that the cascade-forward backpropagation network showed better performance than the feed-forward backpropagation network for predicting the TPC, and the FFBP network, in predicting the DPPH and ABTS antioxidant activity factors, had more precision than the cascade-forward backpropagation network. The ANN technique is a potential method for estimating targeted total phenolic compounds and the antioxidant activity of strawberries.

Mathematical modeling of nutritional, color, texture, and microbial activity changes in fruit and vegetables during drying: A critical review

Retention of quality attributes during drying of fruit and vegetables is a prime concern since the product's acceptability depends on the overall quality; particularly on the nutritional, color, and physical attributes. However, these quality parameters deteriorate during drying. Food quality changes are strongly related to the drying conditions and researchers have attempted to develop mathematical models to understand these relationships. A better insight toward the degradation of quality attributes is crucial for making real predictions and minimizing the quality deterioration. The previous empirical quality models employed kinetic modeling approaches to describe the quality changes and therefore, lack the realistic understanding of fundamental transport mechanisms. In order to develop a physics based mathematical model for the prediction of quality changes during drying, an in-depth understanding of research progress made toward this direction is indispensable. Therefore, the main goal of this paper is to present a critical review of the mathematical models developed and applied to describe the degradation kinetics of nutritional, color, and texture attributes during drying of fruit and vegetables and microbial growth model during storage. This review also presents the advantages and drawbacks of the existing models along with their industrial relevance. Finally, future research propositions toward developing physics-based mathematical model are presented.

Comparison of free, conjugated, and insoluble-bound phenolics and their antioxidant activities in oven-drying and freeze-drying bamboo (Phyllostachys edulis) shoot tips

Bamboo(Phyllostachys edulis) shoot was reported to be rich in phenolics. In the present study, free phenolics, conjugated phenolics, and insoluble-bound phenolics of oven-drying and freeze-drying bamboo shoot tips were extracted and separated, of which total phenolic content (TPC), total flavonoid content (TFC), and their antioxidant activities were determined. Phenolics of different binding forms were qualitatively analyzed using HPLC-ESI-QqQ-MS. A total of 22, 41, and 28 compounds were confirmed or tentatively identified in free, conjugated, and insoluble-bound phenolic extraction, respectively. The majority of the identified compounds were organic acids and phenolic acids. Oven-drying samples exhibited higher TPC (10.53-24.92 mg GAE/100 g DW) and TFC (5.80-33.27 mg CE/100 g DW) values, and stronger antioxidant activities (DPPH, ABTS, and FRAP) than freeze-drying (TPC: 1.67-15.28 mg GAE/100 g DW, TFC: 1.43-29.05 mg CE/100 g DW). Insoluble-bound phenolics were the major contributor to the total antioxidant activity. The present study investigated the phenolics composition and antioxidant activities of different binding forms in bamboo shoot tip comprehensively, and provided available information for their high-value deep-processing.

Application of E-nose technology combined with artificial neural network to predict total bacterial count in milk

Total bacterial count (TBC) is a widely accepted index for assessing microbial quality of milk, and cultivation-based methods are commonly used as standard methods for its measurement. However, these methods are laborious and time-consuming. This study proposes a method combining E-nose technology and artificial neural network for rapid prediction of TBC in milk. The qualitative model generated an accuracy rate of 100% when identifying milk samples with high, medium, or low levels of TBC, on both the testing and validating subsets. Predicted TBC values generated by the quantitative model demonstrated strong coefficient of multiple determination (R² > 0.99) with reference values. Mean relative difference between predicted and reference values (mean ± standard deviation) of TBC were 1.1 ± 1.7% and 0.4 ± 0.8% on the testing and validating subsets involving 24 and 28 tested samples, respectively. Paired t-test implied that the difference between predicted and reference values of TBC was insignificant for both the testing and validating subsets. As low as ∼1 log cfu/mL of TBC present in tested samples were precisely predicted. Results of this study indicated that combination of E-nose technology and artificial neural network generated reliable predictions of TBC in milk. The method proposed in this study was reliable, rapid, and cost efficient for assessing microbial quality milk, and thus would potentially have realistic application in dairy section.

Rapid detection of carmine in black tea with spectrophotometry coupled predictive modelling

Carmine is an artificial colorant commonly used by fraudulent food business participants in black tea adulteration, for purpose of gaining illegal profits. This study combined spectrophotometry with machine learning for rapid detection of carmine in black tea based on the spectral characteristics of tea infusion. The qualitative model demonstrated an accuracy rate of 100% for successful identification of the presence/absence of carmine in black tea. For quantitative analysis, the R² between carmine concentrations generated according to spectral characteristics and those determined with HPLC was 0.988 and 0.972, respectively, for black tea samples involved in the test subset and an independent dataset II. Paired t-test indicated that the difference was statistically insignificant (P values of 0.26 and 0.44, respectively). The method established in this study was rapid and reliable for detecting carmine in black tea, and thus could be used as a useful tool to identify black tea adulteration in market.

Prediction of Pasting Properties of Dough from Mixolab Measurements Using Artificial Neuronal Networks

An artificial neuronal network (ANN) system was conducted to predict the Mixolab parameters which described the wheat flour starch-amylase part (torques C3, C4, C5, and the difference between C3-C4and C5-C4, respectively) from physicochemical properties (wet gluten, gluten deformation index, Falling number, moisture content, water absorption) of 10 different refined wheat flourssupplemented bydifferent levels of fungal α-amylase addition. All Mixolab parameters analyzed and the Falling number values were reduced with the increased level of α-amylase addition. The ANN results accurately predicted the Mixolab parameters based on wheat flours physicochemical properties and α-amylase addition. ANN analyses showed that moisture content was the most sensitive parameter in influencing Mixolab maximum torque C3 and the difference between torques C3 and C4, while wet gluten was the most sensitive parameter in influencing minimum torque C4 and the difference between torques C5 and C4, and α-amylase level was the most sensitive parameter in predicting maximum torque C5. It is obvious that the Falling number of all the Mixolab characteristics best predicted the difference between torques C3 and C4.

Impact of near freezing temperature storage on postharvest quality and antioxidant capacity of two apricot (Prunus armeniaca L.) cultivars

To reduce the postharvest loss and improve apricot quality attributes, near freezing temperature (NFT) technology was applied to store apricot cultivars (var. "Xiaobai" and "Daliguang"). The NFT storage temperatures for the "Xiaobai" apricot and "Daliguang" apricot were determined as -1.9 to -2.3°C and -1.2 to -1.6°C, respectively. Storage at NFT significantly improved the storage quality of apricots by suppressing respiration rate, ethylene production, decay rate, internal browning index, membrane permeability, and malondialdehyde content. Apricots stored at NFT maintained higher firmness, total soluble solids, titrable acid, total phenolics, total flavonoids, and ascorbic acid content than those stored at 0-1°C. Additionally, NFT storage enhanced the capacity of radical scavenging and metal chelating, antioxidant properties in apricots compared to those stored at 0-1°C. Hence, NFT storage proved to be an effective method to improve the quality and antioxidant attributes of apricots. PRACTICAL APPLICATIONS: This study explored the effect of storage at near freezing temperature (NFT) on the postharvest quality of two cultivars of apricot (var. "Xiaobai" and "Daliguang"). We found that storage for 70 days at NFT resulted in better edible quality compared to storage at 0-1°C and 4-6°C. Apricot quality was determined in terms of respiration rate, ethylene production, decay rate, internal browning index, membrane permeability, malondialdehyde content, firmness, total soluble solids, titrable acid, total phenolics, total flavonoids, and ascorbic acid content. The antioxidant properties of the fruits were also retained during storage at NFT. We believe that our study makes a significant contribution to the preservative industry because it demonstrates the superiority of NFT storage over low temperature for apricots.

Bacaba powder produced in spouted bed: an alternative source of bioactive compounds and energy food product

Abstract This study evaluated Bacaba powder produced in a spouted bed as a source of bioactive compounds and high energy value. The conditions influencing the drying process parameters (yield, moisture level, phenolic and anthocyanin retention) as well as simultaneous optimization (optimal conditions) of production were also considered. Drying was most efficient at 75 °C using maltodextrin concentrations above 20.0% (w/w). Higher anthocyanin retention (92.52%) at 65 °C (p = 0.0003), and a maltodextrin concentration of 20.0% (w/w) resulted in high retention of phenolics (95.38%). Accordingly, the operations tested under the desirability function (68 °C, maltodextrin concentration of 21.7% w/w, and air velocity of 1.3 × minimum spouting velocity (Vjm) m s-1) resulted in a process yield of 55.04% and the dry basis (d.b.) composition results were: total phenolics (376.43 mg GAE 100 g-1), energetic value (612.64 kcal 100 g-1), lipids (47.74 g 100 g-1), carbohydrates (27.79 g 100 g-1), protein (15.10 g 100 g-1), and dietetic fiber (8.45 g 100 g-1). The high solubility (92%), flowability (14%), energy, and bioactive characteristics of Bacaba powder suggest the potential for many applications, such as development of dietary supplements, high-energy drinks, milk-based and instant products, and bakery products.

Optimization of ultrasound-assisted extraction of phenolic compounds from grapefruit (Citrus paradisi Macf.) leaves via D-optimal design and artificial neural network design with categorical and quantitative variables

BACKGROUND

The extraction of phenolic compounds from grapefruit leaves assisted by ultrasound-assisted extraction (UAE) was optimized using response surface methodology (RSM) by means of D-optimal experimental design and artificial neural network (ANN). For this purpose, five numerical factors were selected: ethanol concentration (0-50%), extraction time (15-60 min), extraction temperature (25-50 °C), solid:liquid ratio (50-100 g L^-1 ) and calorimetric energy density of ultrasound (0.25-0.50 kW L^-1 ), whereas ultrasound probe horn diameter (13 or 19 mm) was chosen as categorical factor.

RESULTS

The optimized experimental conditions yielded by RSM were: 10.80% for ethanol concentration; 58.52 min for extraction time; 30.37 °C for extraction temperature; 52.33 g L^-1 for solid:liquid ratio; 0.457 kW L^-1 for ultrasonic power density, with thick probe type. Under these conditions total phenolics content was found to be 19.04 mg gallic acid equivalents g^-1 dried leaf.

CONCLUSION

The same dataset was used to train multilayer feed-forward networks using different approaches via MATLAB, with ANN exhibiting superior performance to RSM (differences included categorical factor in one model and higher regression coefficients), while close values were obtained for the extraction variables under study, except for ethanol concentration and extraction time. © 2018 Society of Chemical Industry.

Antioxidant Activity of Selected Polyphenolics in Yeast Cells: The Case Study of Montenegrin Merlot Wine

Screens of antioxidant activity (AA) of various natural products have been a focus of the research community worldwide. This work aimed to differentiate selected samples of Merlot wines originated from Montenegro, with regard to phenolic profile and antioxidant capacity studied by survival rate, total sulfhydryl groups and activities of glutathione peroxidase (GPx), glutathione reductase and catalase in H₂O₂⁻stressed Saccharomyces cerevisiae cells. In this study, DPPH assay was also performed. Higher total phenolic content leads to an enhanced AA under both conditions. The same trend was observed for catechin and gallic acid, the most abundant phenolics in the examined wine samples. Finally, the findings of an Artificial Neural Network (ANN) model were in a good agreement (r² = 0.978) with the experimental data. All tested samples exhibited a protective effect in H₂O₂⁻stressed yeast cells. Pre-treatment with examined wines increased survival in H₂O₂⁻stressed cells and shifted antioxidative defense towards GPx⁻mediated defense. Finally, sensitivity analysis of obtained ANN model highlights the complexity of the impact that variations in the concentrations of specific phenolic components have on the antioxidant defense system.

Recent developments of artificial intelligence in drying of fresh food: A review

Intellectualization is an important direction of drying development and artificial intelligence (AI) technologies have been widely used to solve problems of nonlinear function approximation, pattern detection, data interpretation, optimization, simulation, diagnosis, control, data sorting, clustering, and noise reduction in different food drying technologies due to the advantages of self-learning ability, adaptive ability, strong fault tolerance and high degree robustness to map the nonlinear structures of arbitrarily complex and dynamic phenomena. This article presents a comprehensive review on intelligent drying technologies and their applications. The paper starts with the introduction of basic theoretical knowledge of ANN, fuzzy logic and expert system. Then, we summarize the AI application of modeling, predicting, and optimization of heat and mass transfer, thermodynamic performance parameters, and quality indicators as well as physiochemical properties of dried products in artificial biomimetic technology (electronic nose, computer vision) and different conventional drying technologies. Furthermore, opportunities and limitations of AI technique in drying are also outlined to provide more ideas for researchers in this area.

Interpretation of ANN-based QSAR models for prediction of antioxidant activity of flavonoids

Quantitative structure-activity relationships (QSARs) built using machine learning methods, such as artificial neural networks (ANNs) are powerful in prediction of (antioxidant) activity from quantum mechanical (QM) parameters describing the molecular structure, but are usually not interpretable. This obvious difficulty is one of the most common obstacles in application of ANN-based QSAR models for design of potent antioxidants or elucidating the underlying mechanism. Interpreting the resulting models is often omitted or performed erroneously altogether. In this work, a comprehensive comparative study of six methods (PaD, PaD₂ , weights, stepwise, perturbation and profile) for exploration and interpretation of ANN models built for prediction of Trolox-equivalent antioxidant capacity (TEAC) QM descriptors, is presented. Sum of ranking differences (SRD) was used for ranking of the six methods with respect to the contributions of the calculated QM molecular descriptors toward TEAC. The results show that the PaD, PaD₂ and profile methods are the most stable and give rise to realistic interpretation of the observed correlations. Therefore, they are safely applicable for future interpretations without the opinion of an experienced chemist or bio-analyst. © 2018 Wiley Periodicals, Inc.

Artificial Neural Network Modeling of Drying Kinetics and Color Changes of Ginkgo Biloba Seeds during Microwave Drying Process

Ginkgo biloba seeds were dried in microwave drier under different microwave powers (200, 280, 460, and 640 W) to determinate the drying kinetics and color changes during drying process. Drying curves of all samples showed a long constant rate period and falling rate period along with a short heating period. The effective moisture diffusivities were found to be 3.318 × 10−9 to 1.073 × 10−8 m2/s within the range of microwave output levels and activation energy was 4.111 W/g. The L⁎ and b⁎ values of seeds decreased with drying time. However, a⁎ value decreased firstly and then increased with the increase of drying time. Artificial neural network (ANN) modeling was employed to predict the moisture ratio and color parameters (L⁎, a⁎, and b⁎). The ANN model was trained for finite iteration calculation with Levenberg-Marquardt algorithm as the training function and tansig-purelin as the network transfer function. Results showed that the ANN methodology could precisely predict experimental data with high correlation coefficient (0.9056–0.9834) and low mean square error (0.0014–2.2044). In addition, the established ANN models can be used for online prediction of moisture content and color changes of ginkgo biloba seeds during microwave drying process.