【摘要】 能极大提高肺癌患者生存率的诊断方式是肺癌的早期诊断,基于当前肺癌的早期诊断都依赖于恶性肺结节的检测,而肺结节检测算法数据准备环节最重要的一步是肺实质分割的现状,本文提出一种基于形态学可变元的肺实质分割方法。传统的形态学肺实质分割方法在分割大多数正常肺部CT时,能够达到优良的效果,但是在面对肺部情况复杂,CT层位于肺部顶端或底端时,传统的形态学肺实质分割方法便会失效,此时便需要人工分割。本文提出的这一新方法,能极大提高肺实质分割准确率,摆脱需要人工分割的局面。更多还原

【Abstract】 The diagnosis method that can greatly improve the survival rate of lung cancer patients is the early diagnosis of lung cancer. Based on the current diagnosis of lung cancer, the diagnosis of lung cancer depends on the detection of malignant pulmonary nodules. The most important step in the data preparation of lung nodule detection algorithm is the segmentation of lung parenchyma. We proposed a lung parenchyma segmentation method based on morphological variable elements. The traditional morphological lung parenchymal segmentation method can achieve excellent results when segmenting most normal lung CT, but in the face of complex lung conditions, such as the CT layer is located at the top or bottom of the lung, the traditional morphological lung parenchyma segmentation method will fail and you will need to manually segement it. This new method proposed in this paper can greatly improve the accuracy of segmentation of lung parenchyma and get rid of the situation that requires manual segmentation.更多还原