automatic detection of lung nodules on computer tomography scans with a deep direct regression method

Deep-learning-based approaches have been extensively used in detecting pulmonary nodules from computer tomography (ct) scans. in this work, an automated end-to-end framework with a convolution network (conv-net) is proposed to detect lung nodules from the ct images. here, boundary regression has been performed by a direct regression method, in which the offset is predicted from a given point. the proposed framework has two outputs; a pixel-wise classification between nodule or normal and a direct regression that is used in order to determine the four coordinates of the nodule's bounding box. the loss function includes two terms; one for classification and the other for regression. the performance of the proposed method is compared with yolov2. the evaluation is performed using the lung-pet-ct-dx dataset. the experimental results show that the proposed framework outperforms the yolov2 method. the results obtained demonstrate that the suggested framework possesses high accuracies of nodule localization and boundary estimation