Search

Purpose: Early detection of lung cancer is of importance since it can increase patients’ chances of survival. To detect nodules accurately during screening, radiologists would commonly take the axial, coronal, and sagittal planes into account, rather than solely the axial plane in clinical evaluation. Inspired by clinical work, the paper aims to develop an accurate deep learning framework for nodule detection by a combination of multiple planes. Methods: The nodule detection system is designed in two stages, multiplanar nodule candidate detection, multiscale false positive (FP) reduction. At the first stage, a deeply supervised encoder–decoder network is trained by axial, coronal, and sagittal slices for the candidate detection task. All possible nodule candidates from the three different planes are merged. To further refine results, a three-dimensional multiscale dense convolutional neural network that extracts multiscale contextual information is applied to remove non-nodules. In the public LIDC-IDRI dataset, 888 computed tomography scans with 1186 nodules accepted by at least three of four radiologists are selected to train and evaluate our proposed system via a tenfold cross-validation scheme. The free-response receiver operating characteristic curve is used for performance assessment. Results: The proposed system achieves a sensitivity of 94.2% with 1.0 FP/scan and a sensitivity of 96.0% with 2.0 FPs/scan. Although it is difficult to detect small nodules (i.e.

The research focuses on measuring the influence of V-defect on wedge waves propagating along line wedge tip by using laser ultrasound technique. Generally, wedge has more or less defect or damage on the tip, which may result in break and bring economic losses. Thus it is necessary to investigate characteristic of wedge waves propagating along line wedge with defect. The wedge waveguide models with different defect depth were built by using finite element method. Multiple mode wedge waves were observed through B-scan. The open of defect is 0.1mm, and the depth is 0.01mm, 0.05mm, 0.1mm, 0.2mm, and 0.3mm, respectively. It was seen that both reflected and transmitted waves were observed. Due to the dispersion characteristics, we observed the reflected and transmitted A 1 mode separated from A 2 mode, which can be used to determine the width of V-defect. Meanwhile, models of V-defect with different depth are also built. We had found that wedge waves are totally reflected and there is no transmitted wave observed as the depth is bigger than 0.3mm.

The research focuses on study the influence of truncations on the dispersion of wedge waves propagating along cylinder wedge with different truncations by using the laser ultrasound technique. The wedge waveguide models with different truncations were built by using finite element method (FEM). The dispersion curves were obtained by using 2D Fourier transformation method. Multiple mode wedge waves were observed, which was well agreed with the results estimated from Lagasse’s empirical formula. We established cylinder wedge with radius of 3mm, 20° and 60°angle, with 0μm, 5μm, 10μm, 20μm, 30μm, 40μm, and 50μm truncations, respectively. It was found that non-ideal wedge tip caused abnormal dispersion of the mode of cylinder wedge, the modes of 20° cylinder wedge presents the characteristics of guide waves which propagating along hollow cylinder as the truncation increasing. Meanwhile, the modes of 60° cylinder wedge with truncations appears the characteristics of guide waves propagating along hollow cylinder, and its mode are observed clearly. The study can be used to evaluate and detect wedge structure.