Method for predicting obesity type based on intestinal microorganisms

Abstract

The invention discloses a method for predicting obesity types based on intestinal microorganisms, which comprises the following steps: step 1, collecting data including intestinal flora data of healthy people and intestinal flora data of obese people, and performing data processing aiming at flora diversity; 2, constructing a neural network model, and determining a network structure; step 3, initializing a weight and a deviation; step 4, data propagation; 5, iterative training is carried out; step 6, evaluating and verifying the model; step 7, performing prediction analysis through the model; compared with the prior art, the method has the advantages that flora characteristics and obesity data are continuously iterated through the neural network model, association between different flora of in-vivo intestinal microorganisms and obesity can be obtained, and three-dimensional association between obesity and intestinal microorganisms is constructed; and the human body obesity type can be predicted through the obtained neural network prediction model, and different types of obesity caused by different florae can be obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of intestinal microorganism prediction of obesity, and in particular to a method for predicting obesity type based on intestinal microorganisms. Background Art

[0002] BMI is used as the standard for determining obesity, that is, BMI (Body mass index)>30kg / m' is defined as obesity. Obesity greatly increases the risk of diabetes, hypertension, and atherosclerosis, and the metabolic syndrome caused by this has become the second leading cause of death after smoking. Obesity is very common in developed countries and is becoming increasingly prominent in developing countries. There are more than 500 million obese people in the world, and the number continues to increase. Obesity has become an increasingly serious global public health problem. At present, research on the causes of obesity has progressed rapidly, providing people with a possible explanation for the phenomenon of obesity. It is generally believed that obesity is the result of multiple factors, and the environment, nutrition, personal behavior, genes, etc. are the focus of research. Intestinal microorganisms also have a certain relationship with the formation of obesity.

[0003] Intestinal microorganisms refer to the huge number of microorganisms that exist in the intestines of animals. This group of microorganisms relies on the intestines of animals to live and helps the host complete a variety of physiological and biochemical functions. The intestine is not only an important place for human digestion and absorption, but also the largest immune organ. It plays an extremely important role in maintaining normal immune defense functions. The human intestine provides a good habitat for microorganisms and has metabolic functions that the human body does not have. Intestinal microorganisms help the human body metabolize. By constructing a model of intestinal microorganisms, it is convenient to understand the human intestinal microorganisms.

[0004] The existing association model construction method can only construct models through sample data, and cannot use experimental methods to construct a three-dimensional model between obesity and intestinal microorganisms in the body. In addition, there are many types of intestinal microorganisms, which makes it impossible for the existing association model construction method to prove the association between two or one of the intestinal flora and obesity.

[0005] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgment or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention

[0006] The technical problem to be solved by the present invention is to overcome the above technical defects and provide a method for predicting obesity type based on intestinal microorganisms, which can predict obesity through microbial characteristics and derive the degree of influence of different microbiota on obesity.

[0007] In order to solve the above problems, the technical solution of the present invention is a method for predicting obesity type based on intestinal microorganisms, comprising the following steps:

[0008] Step 1: Collect data, including intestinal flora data of healthy people and intestinal flora data of obese people, and process the data based on flora diversity;

[0009] Step 2: Build a neural network model and determine the network structure;

[0010] Step 3: Initialize weights and biases;

[0011] Step 4: Data dissemination;

[0012] Step 5: Iterative training;

[0013] Step 6: Model evaluation and verification;

[0014] Step 7: Perform predictive analysis through the model.

[0015] Furthermore, in step 1, the healthy population refers to a BMI index between 18.5-25, and the obese population refers to a BMI index greater than 30, and the healthy population and obese population samples are all adults.

[0016] Furthermore, step 2 includes an input layer, a hidden layer, and an output layer. The input layer determines the number of neurons in the input layer according to the number of characteristics of the flora data, determines the number of hidden layers and the number of neurons in each layer, and the output layer determines the number of neurons in the output layer according to the predicted obesity risk.

[0017] Furthermore, obesity types include starch-type obesity, lipid-type obesity, organ-type obesity, metabolic obesity, stubborn obesity, overeating-type obesity, lack of exercise-type obesity, emotional obesity, and physiological genetic obesity.

[0018] Furthermore, in step three, the weights and biases in the network are randomly initialized.

[0019] Furthermore, the data propagation in step 4 includes:

[0020] Forward propagation, the input gut flora data passes through the network, calculates the output of each neuron, and uses the activation function to transform the weighted input of the neuron.

[0021] Loss calculation, calculates the difference between the network output and the actual label, and determines the gradient through the loss function.

[0022] Backpropagation calculates the gradient of the loss with respect to each weight through the chain rule and uses an optimization algorithm to update the weights to reduce the loss.

[0023] Furthermore, in step 5, the processes of forward propagation, loss calculation and back propagation are repeated until the model converges or reaches a predetermined number of training rounds.

[0024] Furthermore, in step six, an independent data set is used to evaluate the model performance to ensure that the model has good generalization ability.

[0025] Furthermore, in step seven, the contribution of different bacterial communities in the model to obesity is analyzed, the trained model is used to predict obesity risk for new samples, and personalized intervention measures are formulated based on the prediction results.

[0026] The advantages of the present invention compared with the prior art are:

[0027] 1. The present invention continuously iterates the data of flora characteristics and obesity through a neural network model, and can conclude that different floras of intestinal microorganisms in the body are associated with obesity, and construct a three-dimensional association between obesity and intestinal microorganisms; and the resulting neural network prediction model can predict the type of human obesity, and it can be concluded that different floras will lead to different types of obesity. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0029] In order to make the contents of the present invention more clearly understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0030] A method for predicting obesity type based on intestinal microorganisms, comprising the following steps:

[0031] Step 1: Collect data, including intestinal flora data of healthy people and intestinal flora data of obese people, and process the data according to flora diversity. The BMI index of healthy people is between 18.5-25, and the BMI index of obese people is greater than 30. The samples of healthy people and obese people are all adults. The types of obesity include starch obesity, lipid obesity, organ obesity, metabolic obesity, stubborn obesity, overeating obesity, lack of exercise obesity, emotional obesity, and physiological genetic obesity.

[0032] Step 2: Construct a neural network model and determine the network structure; it includes an input layer, a hidden layer, and an output layer. The input layer determines the number of neurons in the input layer according to the number of characteristics of different flora data, determines the number of hidden layers and the number of neurons in each layer, and the output layer determines the number of neurons in the output layer according to the predicted obesity risk and the predicted obesity type.

[0033] Step 3: Initialize weights and biases; randomly initialize the weights and biases in the network.

[0034] Step 4: Data dissemination; data dissemination includes:

[0035] Forward propagation, the input gut flora data passes through the network, calculates the output of each neuron, and uses the activation function to transform the weighted input of the neuron.

[0036] Loss calculation, calculates the difference between the network output and the actual label, and determines the gradient through the loss function.

[0037] Backpropagation calculates the gradient of the loss with respect to each weight through the chain rule and uses an optimization algorithm to update the weights to reduce the loss.

[0038] Step 5: Iterative training; repeat the process of forward propagation, loss calculation, and back propagation until the model converges or reaches the predetermined number of training rounds.

[0039] Step 6: Model evaluation and validation: Use independent datasets to evaluate model performance and ensure that the model has good generalization capabilities.

[0040] Step 7: Perform predictive analysis through the model to analyze the contribution of different bacterial communities in the model to different types of obesity, use the trained model to predict obesity risk for new samples, and formulate personalized intervention measures based on the prediction results.

[0041] By continuously iterating the data on microbial flora characteristics and obesity through the neural network model, it can be concluded that different intestinal microbial flora in the body are associated with obesity, and a three-dimensional association between obesity and intestinal microorganisms can be established; and the resulting neural network prediction model can be used to predict the type of human obesity, and it can be concluded that different microbial flora will lead to different types of obesity.

[0042] The present invention and its implementation methods are described above, which is not restrictive. The drawings show only one implementation method of the present invention. In short, if ordinary technicians in the field are inspired by it and design structural methods and embodiments similar to the technical solution without creativity without departing from the purpose of the invention, they should all fall within the protection scope of the present invention.

Claims

1. A method for predicting obesity type based on intestinal microorganisms, characterized in that , including the following steps: Step 1: Collect data, including intestinal flora data of healthy people and intestinal flora data of obese people, and process the data based on flora diversity; Step 2: Build a neural network model and determine the network structure; Step 3: Initialize weights and biases; Step 4: Data dissemination; Step 5: Iterative training; Step 6: Model evaluation and verification; Step 7: Perform predictive analysis through the model.

2. The method for predicting obesity type based on intestinal microorganisms according to claim 1, characterized in that: In step 1, the healthy population refers to a BMI index between 18.5 and 25, and the obese population refers to a BMI index greater than 30. The healthy population and obese population samples are all adults.

3. The method for predicting obesity type based on intestinal microorganisms according to claim 1, characterized in that: The obesity types include starch-type obesity, lipid-type obesity, organ-type obesity, metabolic-type obesity, stubborn-type obesity, overeating-type obesity, lack of exercise-type obesity, emotional-type obesity, and physiological-genetic-type obesity.

4. The method for predicting obesity type based on intestinal microorganisms according to claim 1, characterized in that: The step 2 includes an input layer, a hidden layer, and an output layer. The input layer determines the number of neurons in the input layer according to the number of characteristics of different flora data, determines the number of hidden layers and the number of neurons in each layer, and the output layer determines the number of neurons in the output layer according to the predicted obesity risk and the predicted obesity type.

5. The method for predicting obesity type based on intestinal microorganisms according to claim 1, characterized in that: In step three, the weights and biases in the network are randomly initialized.

6. The method for predicting obesity type based on intestinal microorganisms according to claim 1, characterized in that: The data propagation in step 4 includes: Forward propagation, the input gut flora data passes through the network, calculates the output of each neuron, and uses the activation function to transform the weighted input of the neuron. Loss calculation, calculates the difference between the network output and the actual label, and determines the gradient through the loss function. Backpropagation calculates the gradient of the loss with respect to each weight through the chain rule and uses an optimization algorithm to update the weights to reduce the loss.

7. The method for predicting obesity type based on intestinal microorganisms according to claim 1, characterized in that: In step 5, the process of forward propagation, loss calculation and back propagation is repeated until the model converges or reaches a predetermined number of training rounds.

8. The method for predicting obesity type based on intestinal microorganisms according to claim 1, characterized in that: In step 6, an independent dataset is used to evaluate the model performance to ensure that the model has good generalization ability.

9. The method for predicting obesity type based on intestinal microorganisms according to claim 1, characterized in that: In step seven, the contribution of different bacterial communities in the model to different types of obesity is analyzed, the trained model is used to predict the obesity risk of new samples, and personalized intervention measures are formulated based on the prediction results.

Images