Search for: All records

Interstitial lung disease (ILD) causes pulmonary fibrosis. The correct classification of ILD plays a crucial role in the diagnosis and treatment process. In this research work, we propose a lung nodules recognition method based on a deep convolutional neural network (DCNN) and global features, which can be used for computer-aided diagnosis (CAD) of global features of lung nodules. Firstly, a DCNN is constructed based on the characteristics and complexity of lung computerized tomography (CT) images. Then we discussed the effects of different iterations on the recognition results and influence of different model structures on the global features of lung nodules. We also incorporated the improvement of convolution kernel size, feature dimension, and network depth. Thirdly, the effects of different pooling methods, activation functions and training algorithms we proposed has been analyzed to demonstrate the advantages of the new strategy. Finally, the experimental results verify the feasibility of the proposed DCNN for CAD of global features of lung nodules, and the evaluation shown that our proposed method could achieve an outstanding results compare to state-of-arts. 

Deoxypodophyllotoxin contains a core of four fused rings (A to D) with three consecutive chiral centers, the last being created by the attachment of a peripheral trimethoxyphenyl ring (E) to ring C. Previous studies have suggested that the iron(II)- and 2-oxoglutarate–dependent (Fe/2OG) oxygenase, deoxypodophyllotoxin synthase (DPS), catalyzes the oxidative coupling of ring B and ring E to form ring C and complete the tetracyclic core. Despite recent efforts to deploy DPS in the preparation of deoxypodophyllotoxin analogs, the mechanism underlying the regio- and stereoselectivity of this cyclization event has not been elucidated. Herein, we report 1) two structures of DPS in complex with 2OG and (±)-yatein, 2) in vitro analysis of enzymatic reactivity with substrate analogs, and 3) model reactions addressing DPS’s catalytic mechanism. The results disfavor a prior proposal of on-pathway benzylic hydroxylation. Rather, the DPS-catalyzed cyclization likely proceeds by hydrogen atom abstraction from C7', oxidation of the benzylic radical to a carbocation, Friedel–Crafts-like ring closure, and rearomatization of ring B by C6 deprotonation. This mechanism adds to the known pathways for transformation of the carbon-centered radical in Fe/2OG enzymes and suggests what types of substrate modification are likely tolerable in DPS-catalyzed production of deoxypodophyllotoxin analogs.