A method for rapidly predicting resistant starch content of low glycemic index cereal flour based on near infrared spectroscopy

Abstract

This invention discloses a rapid prediction method for resistant starch content in low glycemic index (GI) cereal flour based on near-infrared spectroscopy, belonging to the field of grain testing technology. This invention collects near-infrared spectra of cereal flour samples treated under different purely physical annealing conditions, with a GI value ≤ 65 and a resistant starch content ≥ 7%. Simultaneously, it uses standard methods to determine the resistant starch content as a reference value and establishes a quantitative prediction model using chemometrics. For the sample to be tested, its near-infrared spectrum is collected and substituted into the model to quickly predict its resistant starch content. This invention requires no chemical reagents, reduces the detection time from tens of hours in traditional methods to just a few minutes, and is specifically optimized for low-GI cereal flour products, suitable for online quality control and product rights protection evidence collection.

Description

Technical Field

[0001] This invention belongs to the field of grain testing technology, specifically relating to a method for rapidly and non-destructively predicting the resistant starch content in low glycemic index cereal flour using near-infrared spectroscopy. Background Technology Resistant starch content is a core indicator for evaluating the quality of low-GI cereal flour. The existing standard method (AOAC 2002.02) requires multiple steps including enzymatic hydrolysis, incubation, centrifugation, washing, and drying, taking approximately 16 hours in total. This cannot meet the needs of online quality control and rapid evidence collection for continuous industrial production. Near-infrared spectroscopy offers advantages such as speed, non-destructive nature, and the absence of chemical reagents; however, existing technologies all rely on predictive models for common starches, and there is currently no specific near-infrared predictive model for resistant starch content in low-GI cereal flours. The applicant has discovered for the first time that low-GI cereal flour modified by pure physical annealing (GI value ≤ 65, resistant starch content ≥ 7%, RS3 type resistant starch melting temperature ≥ 120℃) exhibits a specific crystalline structure and molecular arrangement of resistant starch, which differs from digestible starch and unannealed ordinary starch in the characteristic bands of the near-infrared spectrum. By extracting these characteristic information using chemometric methods, a quantitative prediction model for the resistant starch content of low-GI cereal flour can be established. This method can predict the content of a sample in just a few minutes and does not consume any chemical reagents. Summary of the Invention Purpose of the invention The purpose of this invention is to provide a rapid, non-destructive, and chemical reagent-free method for predicting the resistant starch content of low-GI cereal flour, which can be used for online quality control and product rights protection during the production process of low-GI cereal flour. Technical solution The present invention includes the following steps: (1) collecting near-infrared spectra of low-GI cereal powder after different annealing conditions, and simultaneously determining the resistant starch content using standard methods as a reference value; (2) establishing a prediction model using chemometric methods; (3) collecting near-infrared spectra of the sample to be tested and substituting them into the model to predict its resistant starch content. Core Mechanism Low-GI cereal flour modified by pure physical annealing exhibits resistant starch (RS3 type) with a specific crystalline structure and molecular arrangement, differing from digestible starch and unannealed ordinary starch in the characteristic near-infrared spectral band. By extracting these characteristic information using chemometric methods, a quantitative prediction model specifically for the resistant starch content of low-GI cereal flour can be established. This method can predict the content of a sample in just a few minutes and does not consume any chemical reagents. Beneficial effects (1) Highly targeted: A prediction model is specifically designed for low-GI grain powders (GI≤65, RS≥7%, RS3≥120℃), with higher prediction accuracy than general models. (2) Fast: Reduced from approximately 16 hours to just a few minutes. (3) Non-destructive: Samples can be recycled and reused. (4) No chemical reagents required: Green and environmentally friendly. (5) Applicable to online quality control and patent infringement evidence collection. Detailed Implementation Example 1 (Establishment of a wheat flour prediction model) 120 wheat flour samples were collected and treated under different purely physical annealing conditions (moisture content within the range of 20%-30%, annealing temperature within the range of 50-58℃, annealing time within the range of 8-20 minutes, and relative humidity within the range of 80%-95%). All samples had a GI value ≤65, resistant starch content ≥7%, and RS3 type resistant starch melting temperature ≥120℃. 80 samples were used as the calibration set, and 40 samples as the validation set. Near-infrared spectra (4000-12500 cm⁻¹, diffuse reflectance mode) were collected, and the spectra were preprocessed using standard normal transformation and second derivative. The resistant starch content was determined using the AOAC 2002.02 method as a reference value. A prediction model was established using partial least squares method with 7 principal components. The obtained prediction model calibration set has a determination coefficient R² ≥ 0.90, a cross-validation root mean square error RMSECV ≤ 0.5%, and a small prediction bias in the validation set, which meets the requirements of industrial quality control. Example 2 (Rapid Prediction of Samples to be Tested) An annealed sample of wheat flour with unknown resistant starch content was taken. The sample had a GI value ≤ 65 and an RS3 melting temperature ≥ 120℃. Its near-infrared spectrum was collected and substituted into the model established in Example 1. The predicted value of its resistant starch content was obtained within minutes. The same sample was compared and determined using the AOAC 2002.02 method. The predicted values ​​show a small deviation from those measured by the standard method, meeting industrial quality control requirements. This demonstrates that the method of this invention can accurately and rapidly predict the resistant starch content of low-GI cereal flour. Example 3 (Collection of Evidence for Infringement Determination) Take a commercially available, allegedly infringing product (claiming to have low GI properties), collect its near-infrared spectrum, and substitute it into the prediction model established in Example 1. Within minutes, obtain a predicted value for its resistant starch content. Compare this predicted value with the resistant starch content threshold defined in the relevant patent claims. If the predicted value falls within the protection scope of the relevant patent, it serves as preliminary evidence for infringement determination. Comparative Example 1 (detection using only standard methods) The samples from Example 2 were individually determined using the AOAC 2002.02 enzymatic hydrolysis-gravimetric method. This process involves multiple steps, including enzymatic hydrolysis, incubation, centrifugation, washing, and drying. The entire process took approximately 16 hours, significantly longer than the few minutes required by the method of this invention. This demonstrates that the method of this invention has a significant advantage in detection speed, meeting the timeliness requirements of industrial online quality control and rapid rights protection and evidence collection. Detection methods Near-infrared spectrometer, spectral range 4000-12500 cm⁻¹, diffuse reflectance or transmission mode. Chemometric software employs partial least squares, principal component regression, or artificial neural network methods. Spectral preprocessing methods include standard normal transformation and second derivative calculations. Reference values ​​are determined according to the AOAC 2002.02 standard method.

Claims

1. A rapid prediction method for the resistant starch content of low glycemic index cereal flour based on near-infrared spectroscopy, characterized in that, Includes the following steps: a. Establish a predictive model: Near-infrared spectra of cereal flour samples treated under different purely physical annealing conditions were collected. Simultaneously, the resistant starch content was determined using standard methods as a reference value. A quantitative predictive model between near-infrared spectra and resistant starch content was established using chemometric methods. The cereal flour had a GI value ≤ 65, a resistant starch content ≥ 7%, and an RS3 type resistant starch melting temperature ≥ 120℃. The purely physical annealing conditions included different combinations of parameters selected within the range of moisture content 15%-35%, annealing temperature 50-65℃, annealing time 8-20 minutes, and relative humidity 80%-95%. b. Predict the sample to be tested: Collect the near-infrared spectrum of the cereal flour sample to be tested, and substitute it into the prediction model established in step a to obtain the predicted value of its resistant starch content.

2. The method according to claim 1, characterized in that, The standard method described in step a is AOAC 2002.02 enzymatic hydrolysis-gravimetric method.

3. The method according to claim 1, characterized in that, The chemometric method described in step a is any one of partial least squares, principal component regression, or artificial neural network.

4. The method according to claim 1, characterized in that, The near-infrared spectrum is collected in the range of 4000-12500 cm⁻¹, and the sampling method is diffuse reflection or transmission.

5. The method according to claim 1, characterized in that, The grain flour sample is any one or more of wheat flour, rice flour, corn flour, oat flour, potato flour, and their products modified by pure physical annealing.

6. The method according to claim 1, characterized in that, After the prediction model described in step a is established, the resistant starch content prediction result can be obtained within minutes by collecting the spectrum of the unknown sample and substituting it into the model, without the need for enzymatic hydrolysis, incubation, centrifugation, washing, and drying.

7. The application of the method according to any one of claims 1-6 in online quality control of low-GI cereal flour, characterized in that, Near-infrared spectroscopy acquisition devices are installed on the production line to predict the resistant starch content of each batch of products in real time, thereby achieving full-process quality control.

8. A method for collecting evidence in patent infringement determination, characterized in that, include: The near-infrared spectrum of the allegedly infringing product is collected and substituted into the prediction model described in claim 1 to obtain a predicted value of resistant starch content. This predicted value is then compared with the resistant starch content threshold in the patent claim for reference in determining infringement.

9. The method according to claim 1, characterized in that, The number of samples collected in step a shall not be less than 80, the coefficient of determination of the calibration set R² ≥ 0.90, and the root mean square error of cross-validation RMSECV ≤ 0.5%.

10. A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the method according to any one of claims 1-6.