Radiomics-based brain glioma grading prediction method

Abstract

The present invention provides a method for grading and predicting brain gliomas based on radiomics, comprising the following steps: step 1, collecting original medical images of brain gliomas by conventional magnetic resonance imaging methods, and performing image processing on the collected original medical images Format and size unification and data de-privacy processing, to obtain a glioma medical image sample group; step 2, select one of the glioma medical image samples, and manually segment the sensory information in the glioma medical image sample region of interest; step 3, extracting the features of glioma in the region of interest; step 4, performing dimensionality reduction on the features of glioma extracted in step 3, selecting effective features, and generating a target feature set; step 5, in the brain Sampling with replacement is performed in the glioma medical image sample group, and steps 2 to 4 are repeated to generate a training set; step 6, the prediction model is trained using the Cox regression model according to the training set to generate a prediction model.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine

Login to view more

AI Technical Summary

Problems solved by technology

[0002] The technology based on medical imaging has been applied to the diagnosis and staging of cancer. Due to the characteristics of continuous angiogenesis, tissue invasion and metastasis of glioma, the boundary between normal tissue and glioma tissue is not clear, and the location of metastasis is uncertain. Determining the factors leading to the prognosis prediction of glioma becomes a severe challenge in clinical research

At present,

Images