Improved deep Boltzmann machine-based pulmonary nodule feature extraction and benign and malignant classification method

Abstract

The present invention discloses an improved deep Boltzmann machine-based pulmonary nodule feature extraction and benign and malignant classification method. The method includes the following steps that: step A, pulmonary nodules are segmented from CT images through using a threshold probability image graph method, so that regions of interest (ROI) are obtained, and the regions of interest are cut into nodule images of the same size; and step B, a supervised deep learning algorithm Pnd-EBM is designed to realize the diagnosis of a pulmonary nodule, wherein the diagnosis of the pulmonary nodule further includes three major steps: B1, a deep Boltzmann machine (DBM) is adopted to extract the features of the ROI of the pulmonary nodule which have deep expression abilities; B2, a sparse cross-entropy penalty factor is adopted to improve a cost function, so that the phenomenon of feature homogenization in a training process can be avoided; and B3, an extreme learning machine (ELM) is adopted to perform benign and malignant classification on the extracted features of the pulmonary nodule. The improved deep Boltzmann machine-based pulmonary nodule feature extraction method is superior to a traditional feature extraction method. With the method adopted, the complexity of manual extraction and the difference of feature selection can be avoided, and references can be provided for clinical diagnosis.

Description

technical field [0001] The invention relates to feature extraction of pulmonary nodules, in particular to a method for extracting features of pulmonary nodules and classifying benign and malignant based on an improved deep Boltzmann machine. Background technique [0002] Traditional computer-aided diagnosis (CAD) analysis of pulmonary nodules generally adopts nodule segmentation based on morphology and manual extraction of texture features or shape features of pulmonary nodules. This classification method based on the underlying features can overcome the inertia of the human eye and the insensitivity to grayscale images, and improve the accuracy of nodule recognition and diagnosis by radiologists. To some extent, it depends on experience and luck; and the description of nodules using morphology is inaccurate, such as the definition of nodule margins is vague and subjective. The unsupervised method based on the deep Boltzmann machine can allow the machine to automatically le...

AI Technical Summary

Problems solved by technology

This classification method based on the underlying features can overcome the inertia of the human eye and the insensitivity to grayscale images, and improve the accuracy of nodule recognition and diagnosis by radiologists. Depends on experience and luck to some extent; and the description of nodules using morphology is inaccurate, such as the definition of nodule margins is vague and subjective

Images