Quality analysis and evaluation method for medicinal materials and decoction pieces prone to oil leakage based on electronic nose

Abstract

The invention belongs to the field of quality evaluation of traditional Chinese medicinal materials, and particularly relates to a method for analyzing and evaluating the quality of medicinal materials and decoction pieces easy to get oil based on an electronic nose, which comprises the following steps: preparing a sample, establishing a sensory evaluation scale, determining a quality index and a determination method, determining optimal measurement parameters of the electronic nose, and establishing an oil leakage early warning model. The method comprises the following steps: constructing an oil leakage deterioration early warning model taking smell evaluation as a core; an intelligent and efficient smell comprehensive evaluation method conforming to traditional Chinese medicine characteristics is established, and rapid guidance is provided for traditional Chinese medicine processing, storage and the like.

Description

Technical Field

[0001] The invention belongs to the technical field of quality evaluation of traditional Chinese medicines, and particularly relates to a quality analysis and evaluation method for oil-releasing medicinal materials and decoction pieces based on an electronic nose. Background Art

[0002] Due to the influence of natural storage conditions and processing processes, Chinese herbal medicines and pieces containing more oil often lose oil and go bad. In mild cases, the medicinal properties are weakened, and in severe cases, harmful components may be produced, which seriously affects the safety and effectiveness of clinical medication. At present, the methods for judging whether Chinese herbal medicines and pieces have deteriorated are mainly traditional sensory identification and physical and chemical determination. The former is the most commonly used, but both have certain limitations: sensory identification is highly subjective and relies on human "experience"; the latter is cumbersome, and only measuring one or several components may not fully reflect the quality of Chinese medicine. Rapid identification of the "oil loss" deterioration phenomenon of medicinal materials and pieces, objective judgment of the degree of oil loss, and early warning of oil loss are all urgent issues to be solved, but there are relatively few studies at present.

[0003] Drawing on the methods of chemical medicine and based on modern testing technology, the method of testing one or several components of traditional Chinese medicine is relatively objective, but it also has many shortcomings: the operation is relatively cumbersome and complicated, time-consuming, and relatively expensive, which also requires relatively high technical requirements for operators. Traditional Chinese medicine has complex ingredients, and the method of judging the quality of traditional Chinese medicine only by measuring indicator components is too arbitrary and cannot comprehensively evaluate the quality of traditional Chinese medicine. These undoubtedly bring certain difficulties and limitations to the quality evaluation of traditional Chinese medicine.

[0004] In view of the above problems, the present invention intends to select Ganoderma lucidum spore powder and its broken wall pieces, Prunus armeniaca kernel and its stewed and fried pieces as research carriers, explore the introduction of electronic nose bionic technology in the field of traditional Chinese medicine, reveal the material basis of the changes in property characteristics during the oil loss process; establish an electronic nose detection method and an optimal discrimination model; at the same time, combine the properties and physical and chemical analysis results to comprehensively evaluate the quality of the above-mentioned medicinal materials and pieces, and finally establish an "oil loss" early warning model to provide a reference for how to timely and objectively give early warning values ​​and how to objectively evaluate the quality of traditional Chinese medicines with oil loss. Summary of the invention

[0005] The purpose of the present invention is to provide a quality analysis and evaluation method for oil-releasing medicinal materials and decoction pieces based on an electronic nose to solve the above-mentioned problems.

[0006] The quality analysis and evaluation method of oil-releasing medicinal materials and decoction pieces based on electronic nose includes the following steps:

[0007] Step 1: Sample preparation and establishment of sensory evaluation scale

[0008] Prepare sample materials, and prepare samples with different oil loss degrees based on the long-term sample retention method and the accelerated test method; use the sensory identification method to identify and evaluate the samples, and establish a sensory evaluation scale;

[0009] Step 2: Determine quality indicators and measurement methods

[0010] According to the Chinese Pharmacopoeia, the effective parameters of the samples were determined by high performance liquid chromatography and rancidity determination methods, and the chemical substance data corresponding to the odor changes during the storage and processing of the samples were obtained;

[0011] Step 3: Determine the optimal measurement parameters of the electronic nose

[0012] Taking the maximum response value as the index, the single factor investigation and orthogonal test method were used to investigate the influence of different operating parameters on the response value of the electronic nose and determine the optimal measurement parameters for sample determination.

[0013] Step 4: Establishment of oil leakage warning model

[0014] The principal component analysis and discriminant factor analysis recognition methods were used to conduct electronic nose detection on different samples, analyze the detection results, screen the pattern recognition algorithm, and thus establish the best discriminant model;

[0015] Step 5: Model Validation

[0016] The established sample oil leakage warning model was cross-validated based on the results of sensory evaluation, chemical composition determination and electronic nose analysis.

[0017] A further improvement is that in step 1, the sample material is one or more of Ganoderma lucidum spore powder, Ganoderma lucidum spore broken wall slices, Prunus armeniaca kernel, Prunus armeniaca kernel stewed slices, and Prunus armeniaca kernel fried slices.

[0018] As a further improvement, in step 2, the effective parameters are one or more of amygdalin content, acid value, carbonyl value, triolein, and peroxide value.

[0019] As a further improvement, in step 3, the operating parameters are one or more of sample weight, sample injection volume, particle size, incubation temperature, and incubation time.

[0020] As a further improvement, the algorithm used by the electronic nose is one of a statistical pattern recognition algorithm, a neural network algorithm, a machine learning algorithm and a deep learning algorithm.

[0021] As a further improvement, the algorithm used by the electronic nose is a neural network algorithm, which includes the following steps:

[0022] 1. Data collection:

[0023] Organize and screen the collected sample data;

[0024] 2. Preprocessing and feature extraction:

[0025] The filtered data is preprocessed and feature extracted through normalization and dimensionality reduction to eliminate redundant information and improve computational efficiency;

[0026] Principal component analysis (PCA), independent component analysis (ICA) and other methods are used to reduce the dimension and optimize the data.

[0027] Feature extraction is a key step and can be achieved through methods such as recurrent neural networks (RNNs), convolutional neural networks (CNNs), or deep learning models such as stacked sparse autoencoders (SSAEs).

[0028] 3. Pattern recognition and classification:

[0029] The extracted features are input into the neural network for pattern recognition and classification; commonly used neural networks include back propagation neural network (BPNN), extreme learning machine (ELM), multi-layer perceptron (MLPNN), etc.

[0030] 4. Model training and verification:

[0031] Divide the data set into training set, validation set and test set, and use the training set to train the model;

[0032] Cross-validation, GridSearch and other methods are used to adjust model parameters to prevent overfitting and optimize model performance.

[0033] Use the validation set to evaluate the generalization ability of the model and optimize parameters as needed.

[0034] 5. Result output and application:

[0035] Finally, the model is applied to new samples and the classification results are output.

[0036] The steps of the neural network algorithm of the electronic nose cover the complete process from data collection to feature extraction, pattern recognition, model training and result output. These steps can effectively improve the recognition accuracy and efficiency of the electronic nose system by combining a variety of artificial intelligence technologies and optimization methods.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] 1. The present invention constructs an "oil leakage" deterioration early warning model with "odor evaluation" as the core;

[0039] 2. Establish an intelligent and efficient comprehensive odor evaluation method that conforms to the characteristics of traditional Chinese medicine, and provide quick guidance for the preparation and storage of traditional Chinese medicine. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0041] The present invention is further described in detail below through specific embodiments in conjunction with the accompanying drawings.

[0042] Example 1

[0043] Electronic nose (EN), also known as bionic olfactory system, is an intelligent electronic instrument designed and developed based on the human recognition mechanism of odor, which can measure one or more odor substances. Generally, the electronic nose is mainly composed of three parts: sample processor, gas sensor array, and pattern recognition system. Among them, the gas sensor array and pattern recognition system are the core factors that determine the working efficiency of the electronic nose.

[0044] like Figure 1 As shown:

[0045] 1. Sample preparation and establishment of sensory evaluation scale

[0046] Ganoderma lucidum spore powder and its broken pieces, Prunus armeniaca kernel and its stewed and fried pieces were collected, and samples with different oil loss degrees were prepared based on the long-term sample retention method and accelerated test method. The sensory identification method was used to identify and evaluate the samples, and a sensory evaluation scale was established.

[0047] 2. Determine quality indicators and measurement methods

[0048] In accordance with the "Chinese Pharmacopoeia", high performance liquid chromatography and acidity determination methods were used to determine the amygdalin, acid value, and carbonyl value in mountain peach kernels and their stewed and stir-fried pieces; the triolein and peroxide value of Ganoderma lucidum spore powder and its broken-wall pieces were determined, and the data on chemical substances corresponding to the changes in odor during sample storage and processing were obtained.

[0049] 3. Determine the optimal measurement parameters of the electronic nose

[0050] Taking the maximum response value as the indicator, the single factor investigation and orthogonal test method were used to investigate the effects of multiple factors such as sample weight, injection volume, particle size, incubation temperature, incubation time, etc. on the response value of the electronic nose, and the optimal measurement parameters for the determination of Ganoderma lucidum spore powder and its broken wall pieces, and Prunus armeniaca kernel and its stewed and fried pieces were determined.

[0051] 4. Establishment of oil leakage warning model

[0052] Different pattern recognition methods such as principal component analysis (PCA) and discriminant factor analysis (DFA) were used to carry out electronic nose detection on Ganoderma lucidum spore powder and Prunus armeniaca kernel samples with different processing specifications and different oil loss degrees. The detection results were analyzed and the pattern recognition algorithm was screened to establish the optimal discriminant model.

[0053] 5. Model Validation

[0054] Based on the results of sensory evaluation, chemical composition determination and electronic nose analysis, the established "oil leakage" early warning model for Ganoderma lucidum spore powder, Prunus armeniaca kernel and their decoction pieces was cross-validated.

[0055] Example 2

[0056] Samples of Ganoderma lucidum spore powder and its broken wall pieces, and Prunus armeniaca kernel and its stewed and fried pieces after different processing processes were collected. Based on the constant temperature and humidity accelerated test method, samples with different degrees of "oil loss" were prepared by simulating different temperatures, humidity, light intensities, and different packaging methods. At the same time, a part of the samples were stored in a cool and dry place as specified in the pharmacopoeia for observation.

[0057] The multiple samples collected and prepared as above were subjected to traditional property identification, and grouped according to the identification results, mainly into three groups: "no oil loss", "slight oil loss", and "oil loss". Personnel with professional identification knowledge background were selected to conduct sensory description, and record in detail the color, overall smell, whether there is oil on the surface of each sample, etc., and establish a sensory evaluation form.

[0058] In order to reveal the essential connotation of sensory properties during the "oil loss" process, this experiment used high performance liquid chromatography and acidity determination methods in accordance with the "Chinese Pharmacopoeia" to determine the amygdalin, acid value, and carbonyl value in mountain peach kernels and their stewed and fried pieces, and to determine the triolein and peroxide value of Ganoderma lucidum spore powder and its broken-wall pieces, revealing the basic data of chemical substances that change in odor during sample storage and processing.

[0059] In the identification of Chinese medicine properties, the odor characteristics are directly related to their material basis and are one of the important indicators for evaluating their quality. This experiment first explored and optimized the measurement parameters of the electronic nose to determine the optimal measurement parameters for Ganoderma lucidum spore powder, Prunus armeniaca and their processed products; under this condition, the digital odor extraction of each sample was completed; based on principal component analysis and discriminant factor analysis, the optimal discrimination model of the electronic nose was established; the feasibility and rationality of whether the electronic nose can identify the "lightness" of oil leakage and the "quality" of medicinal materials were verified, that is, whether the odor characteristic fingerprint extracted by the electronic nose is consistent with the traditional identification experience and physical and chemical analysis results, and then explore the potential application value of the electronic nose in the field of Chinese medicine.

[0060] At the same time, the quality is comprehensively evaluated in combination with the results of properties and physical and chemical analysis, and an "oil loss" deterioration early warning model with "odor evaluation" as the core is constructed, thus establishing an intelligent and efficient comprehensive odor evaluation method to provide quick guidance for the storage, maintenance and preparation of Chinese herbal medicines.

[0061] Example 3

[0062] Through the carbonyl value determination method and odor sensory judgment method, the changes in carbonyl values ​​of mountain peach kernels during different stewing, frying and sample observation processes were compared. The material basis for the increase in carbonyl value of "rancid smell" caused by improper operation was found, and the "oil loss" deterioration was reduced by controlling hydrolytic rancidity and oxidative rancidity.

[0063] Example 4

[0064] Through the peroxide value determination method, odor sensory judgment method, and electronic nose bionic technology determination method, the changes in peroxide values ​​of Ganoderma lucidum spore powder and its broken wall pieces with different sample retention times and different processing processes were compared. It was found that the electronic nose response value was positively correlated with the increase in peroxide value of "oil loss" deterioration.

[0065] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A quality analysis and evaluation method for oil-releasing medicinal materials and decoction pieces based on electronic nose, characterized in that: The following steps are involved: Step 1: Sample preparation and establishment of sensory evaluation scale Prepare sample materials, and prepare samples with different oil loss degrees based on long-term sample retention method and accelerated test method; Use sensory identification method to identify and evaluate samples and establish a sensory evaluation scale; Step 2: Determine quality indicators and measurement methods According to the Chinese Pharmacopoeia, the effective parameters of the samples were determined by high performance liquid chromatography and rancidity determination methods, and the chemical substance data corresponding to the odor changes during the storage and processing of the samples were obtained; Step 3: Determine the optimal measurement parameters of the electronic nose Taking the maximum response value as the index, the single factor investigation and orthogonal test method were used to investigate the influence of different operating parameters on the response value of the electronic nose and determine the optimal measurement parameters for sample determination. Step 4: Establishment of oil leakage warning model The principal component analysis and discriminant factor analysis recognition methods were used to conduct electronic nose detection on different samples, analyze the detection results, screen the pattern recognition algorithm, and thus establish the best discriminant model; Step 5: Model Validation The established sample oil leakage warning model was cross-validated based on the results of sensory evaluation, chemical composition determination and electronic nose analysis.

2. The method for quality analysis and evaluation of oil-releasing medicinal materials and decoction pieces based on electronic nose according to claim 1, characterized in that: In step 1, the sample material is one or more of Ganoderma lucidum spore powder, Ganoderma lucidum spore broken wall slices, Prunus armeniaca kernel, Prunus armeniaca kernel stewed slices, and Prunus armeniaca kernel fried slices.

3. The method for quality analysis and evaluation of oil-releasing medicinal materials and decoction pieces based on electronic nose according to claim 1, characterized in that: In step 2, the effective parameters are one or more of amygdalin content, acid value, carbonyl value, triolein, and peroxide value.

4. The method for quality analysis and evaluation of oil-releasing medicinal materials and decoction pieces based on electronic nose according to claim 1, characterized in that: In step 3, the operating parameters are one or more of sample weight, sample injection volume, particle size, incubation temperature, and incubation time.

5. The method for quality analysis and evaluation of oil-releasing medicinal materials and decoction pieces based on electronic nose according to claim 1, characterized in that: The algorithm used by the electronic nose is one of a statistical pattern recognition algorithm, a neural network algorithm, a machine learning algorithm and a deep learning algorithm.

6. The method for quality analysis and evaluation of oil-releasing medicinal materials and decoction pieces based on electronic nose according to claim 5, characterized in that: The algorithm used by the electronic nose is a neural network algorithm, which includes the following steps:

1. Data collection: Organize and screen the collected sample data; 2. Preprocessing and feature extraction: The filtered data is preprocessed and feature extracted through normalization and dimensionality reduction; 3. Pattern recognition and classification: The extracted features are input into a neural network for pattern recognition and classification; 4. Model training and verification: Divide the data set into training set, validation set and test set, and use the training set to train the model; 5. Result output and application: Finally, the model is applied to new samples and the classification results are output.

Images