Background: Osteonecrosis of Femoral Head (ONFH) is a common complication in orthopaedics, wherein femoral structures are usually damaged due to the impairment or interruption head blood supply. Aim: In this study, an automatic approach for classification early ONFH with deep learning has been proposed. Methods: All CT slices according their spatial locations Convolutional Neural Network (CNN) first classified. Therefore, all divided into upper, middle lower segments femur head. Then areas...

To develop an artificial intelligence based three-dimensional (3D) preoperative planning system (AIHIP) for total hip arthroplasty (THA) and verify its accuracy by preliminary clinical application.The CT image database consisting of manually segmented series was built up to train the independently developed deep learning neural network. The network module were assembled within a visual interactive interface-AIHIP. After that, 60 patients (60 hips) with unilateral primary THA between March...

Identifying the right accessories for installing dental implant is a vital element that impacts sustainability and reliability of prosthesis when medical case patient not comprehensive. Dentists need to identify manufacturer from x-ray image determine further treatment procedures. high-pressure task under scaling volume patients pending in queue treatment. To reduce burden on doctors, identification system built based new proposed thinner VGG model with an on-demand client-server structure....

As color and geometry representations are easily recognized by human's brain, data visualization technologies provide mining results with natural intuitive operation interfaces. Currently, users always plot 2D 3D diagrams using the traditional tools. However, it is difficult to visualize big in real time these To enhance quality, this paper proposes a real-time system for data. We apply graphic processing unit (GPU) programming technology algorithms methods parallel. For unified development...

Recently, low‐rank representation (LRR)‐based techniques have manifested remarkable results for data representation. To exploit the latent manifold structure of data, graph regulariser is incorporated into model LRR. However, it critical to construct an appropriate and set corresponding parameters. In addition, this procedure usually time‐consuming proved be overfitting when using cross validation or discrete grid search. Two novel LRR‐based methods, called multiple regularised LRR...

Video content is multifaceted, consisting of objects, scenes, interactions or actions. The existing datasets mostly label only one the facets for model training, resulting in video representation that biases to facet depending on training dataset. There no study yet how learn a from multifaceted labels, and whether information helpful learning. In this paper, we propose new learning framework, MUlti-Faceted Integration (MUFI), aggregate different could reflect full spectrum content....

To summarize the clinical application and research progress in unicompartmental knee arthroplasty (UKA).The literature related to UKA recent years was reviewed emerging indications, implant options, comparisons between other surgical techniques, advances were summarized.Clinical studies show that has many advantages, such as less trauma, faster recovery, fewer postoperative complications. At present, operative indication been expanded. The body mass index more than 25 kg/m 2, 60 old,...

The paper presents the first results of Improved XCS in classification problems. We classify classifiers into certain-right classifiers, certain-wrong and uncertain then analyze difference between certain classifiers. A new classifier attribute at is presented which denotes ascending times prediction error , bad condition set used to improve robustness system. Several experiment on binary integral problems demonstrate superior performance XCS.

In lithology identification method, artificial neural network recognition results du-e to its objective and reliable, be more widely used.Study selection of BP net-work, Shen 630-H1426 logging identification, prediction accuracy 90%, with a higher magmatic rocks metamorphic crocks.Through analysis sho-ws that predict is reliable method.

This paper doing research on the technology of multiple virtual human’s rendering. With problem too much triangular patches needed to be rendering ,make improvement instancing based GPU and combinate method LOD, put forward algorithm large-scale animated human instancing. reduces communicating overhead between CPU GPU, raises efficiency

Video content is multifaceted, consisting of objects, scenes, interactions or actions. The existing datasets mostly label only one the facets for model training, resulting in video representation that biases to facet depending on training dataset. There no study yet how learn a from multifaceted labels, and whether information helpful learning. In this paper, we propose new learning framework, MUlti-Faceted Integration (MUFI), aggregate different could reflect full spectrum content....