To address the problem of automatic identification of fine-grained fracture types, in this paper, we propose a novel framework using 3D convolutional neural network (CNN) to learn fracture features from voxelized bone models which are obtained by establishing isomorphic mapping from fractured bones to a voxelized template. The network, which is named FractureNet, consists of four discriminators forming a multi-stage hierarchy. Each discriminator includes multiple sub-classifiers. These sub-classifiers are chained by two kinds of feature chains (feature map chain and classification feature chain) in the form of a full m-ary tree to perform multi-stage classification tasks. The features learned and classification results obtained at previous stages serve as prior knowledge for current learning and classification. All sub-classifiers are jointly learned in an end-to-end network via a multi-stage loss function integrating losses of the four discriminators. To make our FractureNet more robust and accurate, a data augmentation strategy termed r-combination with constraints is further proposed on the basis of an adjacency relation and a continuity relation between voxels to create a large-scale fracture dataset of voxel models. Extensive experiments show that the proposed method can recognize various fracture types in patients accurately and effectively, and enables significant improvements over the state-of-the-arts on a variety of fracture recognition tasks. Moreover, ancillary experiments on the CIFAR-10 and the PadChest datasets at large scales further support the superior performance of the proposed FractureNet.

Detecting global symmetries in CAD models with free-form surfaces (FFS CAD models) is of great importance in the research of CAD/CAE/CAM integration. A feature-based approach is proposed to rapidly detect global symmetries in FFS CAD models. The key problem is how to detect global symmetries of surface (or curve) models by using feature information. First, our previous basic theories on detecting solid models' symmetries are expanded to handle surface (or curve) models. Then, the feature tuples are used to uniquely describe the surface (or curve) models' non-unique design histories. Based on these studies, the feature reference tree and the parent-child relationships of topological elements are built to construct feature tuples. Finally, the symmetries of the surface (or curve) models are robustly and rapidly detected by using feature information in feature tuples and using the symmetries that have already been detected in solid, surface and curve layers. Experiments show that the FFS CAD models' symmetries can be robustly and rapidly detected by using our approach.

There usually exist narrow-deep areas in mould need to be machined in special machining. To identify the narrow-deep areas automatically, a narrow-deep feature (NF) recognition method in mould is proposed. Firstly, feature hint is extracted from the mould by making use of the characteristic of narrow-deep feature. Then the elementary constituent faces (ECF) of a feature are found starting from the feature hint. The ECF is extended, clipped and the feature faces are obtained incrementally by geometric reasoning. Finally, basic narrow-deep features (BNF) related are combined heuristically. The proposed NF recognition approach provides an intelligent connection between CAD and CAPP for machining narrow-deep areas in mould in special machining.