Detection of idiopathic pulmonary fibrosis lesion regions based on corner point distribution

Idiopathic pulmonary fibrosis (IPF) is one of the most malignant types of interstitial lung disease, severely affecting the quality of life and survival time of patients. Physicians diagnose lesions by reading CT images, but the diverse imaging manifestations of IPF lead to heavy workload and low efficiency. As far as we know, only one automatic IPF detection method is reported. In this paper, we propose a lesion region detection method for IPF, which is divided into two stages, firstly extracting the lesion candidate regions from CT images, and subsequently doing classification by extracting features from the candidate regions to obtain the lesion regions. The experimental results show, the method in this paper can accurately detect the lesion regions and effectively improve the detection accuracy compared with the k-means clustering-based IPF detection method.