The Experts below are selected from a list of 308754 Experts worldwide ranked by ideXlab platform

Jan Egger - One of the best experts on this subject based on the ideXlab platform.

  • a baseline approach for autoimplant the miccai 2020 cranial implant Design Challenge
    2020 Workshop on Clinical Image-Based Procedures: in Conjunction with MICCAI 2020, 2020
    Co-Authors: Antonio Pepe, Christina Gsaxner, Gord Von Campe, Jan Egger
    Abstract:

    In this study, we present a baseline approach for AutoImplant (https://autoimplant.grand-Challenge.org/) – the cranial implant Design Challenge, which can be formulated as a volumetric shape learning task. In this task, the defective skull, the complete skull and the cranial implant are represented as binary voxel grids. To accomplish this task, the implant can be either reconstructed directly from the defective skull or obtained by taking the difference between a defective skull and a complete skull. In the latter case, a complete skull has to be reconstructed given a defective skull, which defines a volumetric shape completion problem. Our baseline approach for this task is based on the former formulation, i.e., a deep neural network is trained to predict the implants directly from the defective skulls. The approach generates high-quality implants in two steps: First, an encoder-decoder network learns a coarse representation of the implant from downsampled, defective skulls; The coarse implant is only used to generate the bounding box of the defected region in the original high-resolution skull. Second, another encoder-decoder network is trained to generate a fine implant from the bounded area. On the test set, the proposed approach achieves an average dice similarity score (DSC) of 0.8555 and Hausdorff distance (HD) of 5.1825 mm. The codes are available at https://github.com/Jianningli/autoimplant.

  • A Baseline Approach for AutoImplant: the MICCAI 2020 Cranial Implant Design Challenge
    arXiv: Computer Vision and Pattern Recognition, 2020
    Co-Authors: Antonio Pepe, Christina Gsaxner, Gord Von Campe, Jan Egger
    Abstract:

    In this study, we present a baseline approach for AutoImplant (this https URL) - the cranial implant Design Challenge, which, as suggested by the organizers, can be formulated as a volumetric shape learning task. In this task, the defective skull, the complete skull and the cranial implant are represented as binary voxel grids. To accomplish this task, the implant can be either reconstructed directly from the defective skull or obtained by taking the difference between a defective skull and a complete skull. In the latter case, a complete skull has to be reconstructed given a defective skull, which defines a volumetric shape completion problem. Our baseline approach for this task is based on the former formulation, i.e., a deep neural network is trained to predict the implants directly from the defective skulls. The approach generates high-quality implants in two steps: First, an encoder-decoder network learns a coarse representation of the implant from down-sampled, defective skulls; The coarse implant is only used to generate the bounding box of the defected region in the original high-resolution skull. Second, another encoder-decoder network is trained to generate a fine implant from the bounded area. On the test set, the proposed approach achieves an average dice similarity score (DSC) of 0.8555 and Hausdorff distance (HD) of 5.1825 mm. The code is publicly available at this https URL.

Krishnendu Chakrabarty - One of the best experts on this subject based on the ideXlab platform.

  • unified high level synthesis and module placement for defect tolerant microfluidic biochips
    Design Automation Conference, 2005
    Co-Authors: Fei Su, Krishnendu Chakrabarty
    Abstract:

    Microfluidic biochips promise to revolutionize biosensing and clinical diagnostics. As more bioassays are executed concurrently on a biochip, system integration and Design complexity are expected to increase dramatically. This problem is also identified by the 2003 ITRS document as a major system-level Design Challenge beyond 2009. We focus here on the automated Design of droplet-based microfluidic biochips. We present a synthesis methodology that unifies operation scheduling, resource binding, and module placement for such "digital" biochips. The proposed technique, which is based on parallel recombinative simulated annealing, can also be used after fabrication to bypass defective cells in the microfluidic array. A real-life protein assay is used to evaluate the synthesis methodology.

Grant Martin - One of the best experts on this subject based on the ideXlab platform.

  • overview of the mpsoc Design Challenge
    Design Automation Conference, 2006
    Co-Authors: Grant Martin
    Abstract:

    We review the Design Challenges faced by MPSoC Designers at all levels. Starting at the application level, there is a need for programming models and communications APIs that allow applications to be easily re-configured for many different possible architectures without tedious rewriting, while at the same time ensuring efficient production code. Synchronisation and control of task scheduling may be provided by RTOS's or other scheduling methods, and the choice of programming and threading models, whether symmetric or asymmetric, has a heavy influence on how best to control task or thread execution. Debugging MP systems for the typical application developer becomes a much more complex job, when compared to traditional single-processor debug, or the debug of simple MP systems that are only very loosely coupled. The interaction between the system, applications and software views, and processor configuration and extension, adds a new dimension to the problem space. Zeroing in on the optimal solution for a particular MPSoC Design demands a multi-disciplinary approach. After reviewing the Design Challenges, we end by focusing on the requirements for Design tools that may ameliorate many of these issues, and illustrate some of the possible solutions, based on experiments.

Antonio Pepe - One of the best experts on this subject based on the ideXlab platform.

  • a baseline approach for autoimplant the miccai 2020 cranial implant Design Challenge
    2020 Workshop on Clinical Image-Based Procedures: in Conjunction with MICCAI 2020, 2020
    Co-Authors: Antonio Pepe, Christina Gsaxner, Gord Von Campe, Jan Egger
    Abstract:

    In this study, we present a baseline approach for AutoImplant (https://autoimplant.grand-Challenge.org/) – the cranial implant Design Challenge, which can be formulated as a volumetric shape learning task. In this task, the defective skull, the complete skull and the cranial implant are represented as binary voxel grids. To accomplish this task, the implant can be either reconstructed directly from the defective skull or obtained by taking the difference between a defective skull and a complete skull. In the latter case, a complete skull has to be reconstructed given a defective skull, which defines a volumetric shape completion problem. Our baseline approach for this task is based on the former formulation, i.e., a deep neural network is trained to predict the implants directly from the defective skulls. The approach generates high-quality implants in two steps: First, an encoder-decoder network learns a coarse representation of the implant from downsampled, defective skulls; The coarse implant is only used to generate the bounding box of the defected region in the original high-resolution skull. Second, another encoder-decoder network is trained to generate a fine implant from the bounded area. On the test set, the proposed approach achieves an average dice similarity score (DSC) of 0.8555 and Hausdorff distance (HD) of 5.1825 mm. The codes are available at https://github.com/Jianningli/autoimplant.

  • A Baseline Approach for AutoImplant: the MICCAI 2020 Cranial Implant Design Challenge
    arXiv: Computer Vision and Pattern Recognition, 2020
    Co-Authors: Antonio Pepe, Christina Gsaxner, Gord Von Campe, Jan Egger
    Abstract:

    In this study, we present a baseline approach for AutoImplant (this https URL) - the cranial implant Design Challenge, which, as suggested by the organizers, can be formulated as a volumetric shape learning task. In this task, the defective skull, the complete skull and the cranial implant are represented as binary voxel grids. To accomplish this task, the implant can be either reconstructed directly from the defective skull or obtained by taking the difference between a defective skull and a complete skull. In the latter case, a complete skull has to be reconstructed given a defective skull, which defines a volumetric shape completion problem. Our baseline approach for this task is based on the former formulation, i.e., a deep neural network is trained to predict the implants directly from the defective skulls. The approach generates high-quality implants in two steps: First, an encoder-decoder network learns a coarse representation of the implant from down-sampled, defective skulls; The coarse implant is only used to generate the bounding box of the defected region in the original high-resolution skull. Second, another encoder-decoder network is trained to generate a fine implant from the bounded area. On the test set, the proposed approach achieves an average dice similarity score (DSC) of 0.8555 and Hausdorff distance (HD) of 5.1825 mm. The code is publicly available at this https URL.

Fei Su - One of the best experts on this subject based on the ideXlab platform.

  • unified high level synthesis and module placement for defect tolerant microfluidic biochips
    Design Automation Conference, 2005
    Co-Authors: Fei Su, Krishnendu Chakrabarty
    Abstract:

    Microfluidic biochips promise to revolutionize biosensing and clinical diagnostics. As more bioassays are executed concurrently on a biochip, system integration and Design complexity are expected to increase dramatically. This problem is also identified by the 2003 ITRS document as a major system-level Design Challenge beyond 2009. We focus here on the automated Design of droplet-based microfluidic biochips. We present a synthesis methodology that unifies operation scheduling, resource binding, and module placement for such "digital" biochips. The proposed technique, which is based on parallel recombinative simulated annealing, can also be used after fabrication to bypass defective cells in the microfluidic array. A real-life protein assay is used to evaluate the synthesis methodology.