Method and device for determining lung nodules to be benign or malignant

Abstract

The invention discloses a method and a device for determining lung nodules to be benign or malignant. The method comprises the following steps of extracting a lung nodule image from a lung CT image ofa patient, carrying out feature extraction by adopting a preset feature extracting neural network model, and obtaining a feature vector corresponding to the patient; further, inputting the feature vector corresponding to the patient into a preset benign and malignant classification neural network model, and obtaining benign or malignant properties of lung modules suffered by the patient from thepreset benign and malignant classification neural network model. Accordingly, compared with the mode that a doctor determines the lung nodules to be benign or malignant through human diagnosis in theprior art, the invention is advantageous in that the preset feature extracting neural network model is adopted, and the benign and malignant lung nodules are determined by combining with a preset benign and malignant classification neural network; the diagnosis error rate caused by the level difference of doctors can be reduced, and the accuracy of determining the benign and malignant lung nodulesis improved.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a method and device for determining benign and malignant pulmonary nodules. Background technique [0002] A pulmonary nodule is a noncaseating necrotic epithelial granulomatous lesion. Pulmonary nodules can be divided into benign and malignant. Benign pulmonary nodules can be relieved on their own, and the harm to the human body is relatively small; malignant pulmonary nodules can easily cause functional damage, extensive lung fibrosis, and even lung cancer, which is harmful to the human body. great. Therefore, when a patient suffers from pulmonary nodules, it is necessary to accurately determine whether the pulmonary nodules suffered by the patient are benign pulmonary nodules or malignant pulmonary nodules. [0003] In the prior art, when doctors determine benign and malignant pulmonary nodules, they often observe chest images. This diagnostic method requires doctors ...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, when doctors determine benign and malignant pulmonary nodules, they often observe chest images. This diagnostic method requires doctors to make judgments manually, which is inefficient and is easily affected by regional medical level and personal experience level of doctors. impact, less accurate

Images