Method for automatically evaluating difficulty of kidney tumor enucleation based on CT (Computed Tomography) image

Abstract

The invention discloses a method for automatically evaluating difficulty of kidney tumor enucleation based on CT (computed tomography) images, which comprises the following steps: based on artificial intelligence and image analysis technologies, segmenting an abdominal wall, a kidney and a kidney tumor at the same time by establishing a deep network model, automatically calculating an MAP score and an R.E.N.A.L score by utilizing relevance among the tumor, a renal pelvis, the kidney and the abdominal wall, and automatically evaluating the difficulty of the kidney tumor enucleation based on the CT images. Therefore, the difficulty of the kidney tumor enucleation is reasonably and effectively evaluated; the method has the advantages of time and labor saving and high evaluation accuracy and reliability.

Description

technical field [0001] The invention relates to a method for evaluating the difficulty of enucleation of renal tumors, in particular to an automatic evaluation method for the difficulty of enucleation of renal tumors based on CT images. Background technique [0002] Early-stage renal cancer can be effectively treated with renal tumor enucleation. Although the quality of instruments such as laparoscope or Da Vinci robot is constantly improving and improving, and surgeons are constantly skilled, the difficulty of surgery caused by individual differences of patients cannot be ignored. In general, the difficulty of surgery related to individual differences of patients mainly lies in two aspects. The first aspect is the thickness of the perirenal fat and the degree of adhesion; the second aspect is the size of the tumor and the location of the tumor in the kidney. The first aspect increases the difficulty of freeing the kidney during the operation, and the second aspect determin...

AI Technical Summary

Problems solved by technology

[0003] However, the above two scores only provide the scoring criteria for the difficulty of enucleation. In current practical applications, doctors need to subjectively analyze the CT images, and then compare the two scoring criteria to give the corresponding difficulty score.

Therefore, the existing methods for assessing the difficulty of renal tumor enucleation are based on manual observation of CT images, which is time-consuming and laborious, and is affected by the subjectivity of doctors, making it difficult to guarantee the accuracy and reliability of the assessment.

Images