Gastrointestinal stromal tumor prediction method and system based on XGBoost algorithm

Abstract

The invention discloses a gastrointestinal stromal tumor prediction method and a gastrointestinal stromal tumor prediction system based on an XGBoost algorithm, and the method comprises the following steps: obtaining examination data most related to gastrointestinal stromal tumor diagnosis from enhanced CT (Computed Tomography), an ultrasonic endoscope and blood test indexes during preliminary diagnosis of cases which are preliminarily diagnosed as cases with gastrointestinal stromal tumor; dividing the examination data into a training data set and a verification data set; selecting model parameters and constructing an XGBoost prediction model by using the model parameters and the training data set; verifying the XGBoost prediction model by using the verification data set; and inputting data of a to-be-predicted patient which is preliminarily diagnosed as a patients with gastrointestinal stromal tumor into the XGBoost prediction model, and outputting a prediction result through model calculation. According to the method, all factors possibly influencing gastrointestinal stromal tumor diagnosis are included, preoperative prediction is carried out on the tumor of the patient comprehensively, the prediction accuracy is high, the economic burden of the patient is reduced, and extra pain does not need to be suffered.

Description

technical field

[0001] The present invention relates to the technical field of medical data processing, more specifically, to a method and system for predicting gastrointestinal stromal tumors based on XGBoost algorithm. Background technique

[0002] Gastrointestinal stromal tumor (GIST) is currently the most common soft tissue tumor of the gastrointestinal tract, mainly occurring in the stomach, followed by the small intestine, colon and other tissues outside the gastrointestinal tract. Because GIST has the biological behavior of malignant tumors, the best treatment recommended by the guidelines is complete surgical resection, but most other soft tissue tumors in the gastrointestinal tract are not highly malignant and can be treated without surgery. Therefore, it is particularly important for clinicians to accurately identify GIST before surgery.

[0003] The most accurate method for preoperative diagnosis of GIST is fine-needle aspiration biopsy. However, because this tec...

Examples

example 1

[0099] Create a data frame and input the above six indicators of patient A to be predicted. For example, the ratio of the long and short diameters of his input tumor is 1.5454545, the plain scan CT value of the tumor is 12, the tumor is uniformly enhanced under enhanced CT, and endoscopic ultrasonography shows that there is no ulcer on the surface of the tumor and there is a liquid dark area inside. The peripheral lymphocyte count was 1.56 (×10 9 / L). As follows: datanewpatient<-data.frame(Long.Short.Diameter=1.5454545, CT.Value=12, Homogeneously.Enhanced=1, Ulcer=0, Liquid.Area=1, Lymphcte.Count=1.56)

[0100] Next, input the above data into the XGBoost prediction model and adjust the data format. After the calculation of the model, the final prediction result is output. like image 3 As shown, it can be seen that the calculated patient prediction value is 0.373, which is smaller than the measured value 0.666 (intercept value) predicted by the model, so the model output r...

example 2

[0102] Create a data frame and input the above six indicators of patient B to be predicted. For example, the ratio of the long and short diameters of his tumor is 1.053, the plain scan CT value of the tumor is 33, the tumor is unevenly enhanced under enhanced CT, and endoscopic ultrasonography shows that there is no ulcer on the surface of the tumor and there is a liquid dark area inside. The peripheral lymphocyte count was 1.7 (×10 9 / L). The following table: datanewpatient<-data.frame(Long.Short.Diameter=1.053, CT.Value=33, Homogeneously.Enhanced=0, Ulcer=0, Liquid.Area=1, Lymphcte.Count=1.7)

[0103] Next, input the above data into the XGBoost prediction model and adjust the data format. After the calculation of the model, the final prediction result is output. like Figure 4 As shown, it can be seen that the calculated patient prediction value is 0.898, which is larger than the model prediction measurement value of 0.666 (intercept value), so the model output result is...

technical field

[0001] The present invention relates to the technical field of medical data processing, more specifically, to a method and system for predicting gastrointestinal stromal tumors based on XGBoost algorithm. Background technique

[0002] Gastrointestinal stromal tumor (GIST) is currently the most common soft tissue tumor of the gastrointestinal tract, mainly occurring in the stomach, followed by the small intestine, colon and other tissues outside the gastrointestinal tract. Because GIST has the biological behavior of malignant tumors, the best treatment recommended by the guidelines is complete surgical resection, but most other soft tissue tumors in the gastrointestinal tract are not highly malignant and can be treated without surgery. Therefore, it is particularly important for clinicians to accurately identify GIST before surgery.

[0003] The most accurate method for preoperative diagnosis of GIST is fine-needle aspiration biopsy. However, because this tec...

Examples

example 1

[0099] Create a data frame and input the above six indicators of patient A to be predicted. For example, the ratio of the long and short diameters of his input tumor is 1.5454545, the plain scan CT value of the tumor is 12, the tumor is uniformly enhanced under enhanced CT, and endoscopic ultrasonography shows that there is no ulcer on the surface of the tumor and there is a liquid dark area inside. The peripheral lymphocyte count was 1.56 (×10 9 / L). As follows: datanewpatient<-data.frame(Long.Short.Diameter=1.5454545, CT.Value=12, Homogeneously.Enhanced=1, Ulcer=0, Liquid.Area=1, Lymphcte.Count=1.56)

[0100] Next, input the above data into the XGBoost prediction model and adjust the data format. After the calculation of the model, the final prediction result is output. like image 3 As shown, it can be seen that the calculated patient prediction value is 0.373, which is smaller than the measured value 0.666 (intercept value) predicted by the model, so the model output r...

example 2

[0102] Create a data frame and input the above six indicators of patient B to be predicted. For example, the ratio of the long and short diameters of his tumor is 1.053, the plain scan CT value of the tumor is 33, the tumor is unevenly enhanced under enhanced CT, and endoscopic ultrasonography shows that there is no ulcer on the surface of the tumor and there is a liquid dark area inside. The peripheral lymphocyte count was 1.7 (×10 9 / L). The following table: datanewpatient<-data.frame(Long.Short.Diameter=1.053, CT.Value=33, Homogeneously.Enhanced=0, Ulcer=0, Liquid.Area=1, Lymphcte.Count=1.7)

[0103] Next, input the above data into the XGBoost prediction model and adjust the data format. After the calculation of the model, the final prediction result is output. like Figure 4 As shown, it can be seen that the calculated patient prediction value is 0.898, which is larger than the model prediction measurement value of 0.666 (intercept value), so the model output result is...

Images