engleski

Database with Annotated Lung CT Images

Lung CT (Computed Tomography) is a diagnostic procedure essential for detecting a range of lesions and diseases. Different machine learning algorithms have been applied for the detection and classification of abnormalities. For training and testing the algorithms it is necessary to have a database with a range of abnormalities in the lungs. We propose a novel database ALCTID (Annotated Lung CT Image Database) in which an experienced thoracic radiologist manually annotated regions of interest (ROI) with abnormalities. Database includes 170 images of healthy lungs and 170 lung CT images with a total of 307 annotated ROIs comprising a range of abnormalities. The database is publicly available and suitable for use by students learning about the application of deep learning and computer-aided diagnosis for medical image processing. Finally, we applied a convolutional neural network (CNN) to demonstrate the usability of the database in the automatic detection of changes in the lung tissue.

Lung ; CT ; Database ; Region of Interest ; Neural Network ; ALCTID

nije evidentirano