Imaging omics feature processing method and device for flat scanning CT image

Abstract

The invention provides an imaging omics feature processing method and device for a flat scanning CT image. The method comprises the steps: obtaining a chest flat-scanning thin-layer CT image which comprises a CT image of novel coronavirus pneumonia and CT images of other viral pneumonia in the same period; carrying out region-of-interest sketching on the CT image layer by layer, and extracting animage omics feature in the region; modeling by using the gradient boosting tree model, determining model parameters of the gradient boosting tree model based on an improved genetic algorithm, and performing feature selection to obtain final model parameters and a final feature selection result, wherein the model parameters comprise the maximum weak learner number, the learning rate and the sub-sampling proportion; constructing a final gradient boosting tree model by using the final feature selection result and the final model parameters; adding the probability of predicting the novel coronavirus pneumonia of each decision tree by using the final gradient boosting tree model to obtain a final iconography marker feature.

Description

technical field [0001] The invention relates to the field of computers, in particular to a radiomics feature processing method and device for plain scan CT images. Background technique [0002] In recent years, radiomics has been widely used in many areas of cancer, such as tumor detection, preoperative lymph node metastasis prediction, and treatment response assessment, due to its ability to extract high-throughput quantitative features from conventional CT images. Radiomics research generally includes data collection, region-of-interest delineation, feature extraction, feature screening, and modeling. The extracted features should comply with the Imaging Biomarker Standardization Initiative (IBSI). [0003] The novel coronavirus (COVID-19) is highly contagious and can cause severe acute dyspnea or multiple organ failure in severe cases. The World Health Organization (WHO) declared the COVID-19 outbreak a "Public Health Emergency of International Concern" (PHEIC) on Janua...

AI Technical Summary

Problems solved by technology

However, COVID-19 may have some similar CT imaging features from other types of viral pneumonia, making it difficult to distinguish

Images