Abstract
ObjectivesAnterior mediastinal disease is a common disease in the chest. Computed tomography (CT), as an important imaging technology, is widely used in the diagnosis of mediastinal diseases. Doctors find it difficult to distinguish lesions in CT images because of image artifact, intensity inhomogeneity, and their similarity with other tissues. Direct segmentation of lesions can provide doctors a method to better subtract the features of the lesions, thereby improving the accuracy of diagnosis.MethodAs the trend of image processing technology, deep learning is more accurate in image segmentation than traditional methods. We employ a two-stage 3D ResUNet network combined with lung segmentation to segment CT images. Given that the mediastinum is between the two lungs, the original image is clipped through the lung mask to remove some noises that may affect the segmentation of the lesion. To capture the feature of the lesions, we design a two-stage network structure. In the first stage, the features of the lesion are learned from the low-resolution downsampled image, and the segmentation results under a rough scale are obtained. The results are concatenated with the original image and encoded into the second stage to capture more accurate segmentation information from the image. In addition, attention gates are introduced in the upsampling of the network, and these gates can focus on the lesion and play a role in filtering the features. The proposed method has achieved good results in the segmentation of the anterior mediastinal.ResultsThe proposed method was verified on 230 patients, and the anterior mediastinal lesions were well segmented. The average Dice coefficient reached 87.73%. Compared with the model without lung segmentation, the model with lung segmentation greatly improved the accuracy of lesion segmentation by approximately 9%. The addition of attention gates slightly improved the segmentation accuracy.ConclusionThe proposed automatic segmentation method has achieved good results in clinical data. In clinical application, automatic segmentation of lesions can assist doctors in the diagnosis of diseases and may facilitate the automated diagnosis of illnesses in the future.
Highlights
Given the high incidence of chest diseases, the anterior mediastinal disease is an urgent medical condition
A total of 116 cases were used as the training data and 114 cases as the test data to distinguish the differences in the data caused by the imaging equipment and to better highlight the advantages of deep learning models
This study shows that deep learning can achieve good results in the segmentation of anterior mediastinal lesions and provides a relatively reliable basis for subsequent clinical diagnosis
Summary
Given the high incidence of chest diseases, the anterior mediastinal disease is an urgent medical condition. The anterior mediastinal lesion is characterized by low contrast, irregular shape, different sizes, and unstable anatomical positions These features lead to challenges to the image acquisition and analysis of anterior mediastinal disease and difficulties for doctors to make diagnosis. Some traditional segmentation methods need manual interaction, such as region growth [14] and graphcut [15, 16] Some conventional techniques, such as snakes [17, 18] and active contour model [19, 20], require the manual setting of many parameters. These processes cannot achieve fully automatic results, and the segmentation results for low-contrast medical images are poor
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
