Browsing by Subject "Three-dimensional displays"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    3D Facial Matching by Spiral Convolutional Metric Learning and a Biometric Fusion-Net of Demographic Properties
    (IEEE, 2021) Mahdi, Soha Sadat; Nauwelaers, Nele; Joris, Philip; Bouritsas, Giorgos; Gong, Shunwang; Bokhnyak, Sergiy; Walsh, Susan; Shriver, Mark D.; Bronstein, Michael; Claes, Peter; Biology, School of Science
    Face recognition is a widely accepted biometric verification tool, as the face contains a lot of information about the identity of a person. In this study, a 2-step neural-based pipeline is presented for matching 3D facial shape to multiple DNA-related properties (sex, age, BMI and genomic background). The first step consists of a triplet loss-based metric learner that compresses facial shape into a lower dimensional embedding while preserving information about the property of interest. Most studies in the field of metric learning have only focused on 2D Euclidean data. In this work, geometric deep learning is employed to learn directly from 3D facial meshes. To this end, spiral convolutions are used along with a novel mesh-sampling scheme that retains uniformly sampled 3D points at different levels of resolution. The second step is a multi-biometric fusion by a fully connected neural network. The network takes an ensemble of embeddings and property labels as input and returns genuine and imposter scores. Since embeddings are accepted as an input, there is no need to train classifiers for the different properties and available data can be used more efficiently. Results obtained by a to-fold cross-validation for biometric verification show that combining multiple properties leads to stronger biometric systems. Furthermore, the proposed neural-based pipeline outperforms a linear baseline, which consists of principal component analysis, followed by classification with linear support vector machines and a Naïve Bayes-based score-fuser.
  • Thumbnail Image
    Item
    An Open Source Platform for Computational Histopathology
    (IEEE, 2021) Yu, Xiaxia; Zhao, Bingshuai; Huang, Haofan; Tian, Mu; Zhang, Sai; Song, Hongping; Li, Zengshan; Huang, Kun; Gao, Yi; Biostatistics and Health Data Science, School of Medicine
    Computational histopathology is a fast emerging field which converts the traditional glass slide based department to a new examination platform. Such a paradigm shift also brings the in silico computation to the field. Much research have been presented in the past decades on the algorithm development for pathology image analysis. On the other hand, a comprehensive software platform with advanced visualization and computation capability, large developer community, flexible plugin mechanism, and friendly transnational license, would be extremely beneficial for the entire community. In this work, we present SlicerScope: an open platform for whole slide histopathology image computing based on the highly successful 3D Slicer. We present rationale on the choice of such an architecture, introducing new modules/tools for giga-pixel whole slide image viewing, and four specific analytical modules for qualitative presentation, nucleus level analysis, tissue scale computation, and 3D pathology. The entire software is publicly available at https://slicerscope.github.io/ , facilitating the algorithmic, clinical, and transnational researches.
  • Thumbnail Image
    Item
    Direct and Extended Piezoresistive and Piezoelectric Strain Fusion for a Wide Band PVDF/MWCNT-Based 3D Force Sensor
    (IEEE, 2021) Alotaibi, Ahmed; Anwar, Sohel; Mechanical and Energy Engineering, School of Engineering and Technology
    This paper presents a novel 3D force sensor design based on in-situ nanocomposite strain sensors. The polymer matrix of the polyvinylidene fluoride (PVDF) and multi-walled carbon nanotubes (MWCNT) conductive filler nanocomposite film have been chosen as sensing elements for the 3D force sensor. A bioinspired tree branch design was used as the 3D force sensor’s elastic structure, that was built using thin Aluminum plates and a laser cutting fabrication process. The PVDF/MWCNT films contained piezoresistive and piezoelectric characteristics, allowing for static/low and dynamic strain measurements, respectively. Two compositions with 0.1 and 2 wt.% PVDF/MWCNT sensing elements were selected for piezoelectric and piezoresistive strain measurements, respectively. These characteristic measurements were investigated under different loading frequencies in a simply supported beam experiment. The 3D force sensor was tested under dynamic excitation in the Z-direction and the X-direction. A Direct Piezoresistive/Piezoelectric fusion (DPPF) method was developed by fusing the piezoresistive and piezoelectric measurements at a given frequency that overcomes the limited frequency ranges of each of the strain sensor characteristics. The DPPF method is based on a fuzzy inference system (FIS) which is constructed and tuned using the subtractive clustering technique. Different nonlinear Hammerstein-Wiener (nlhw) models were used to estimate the actual strain from piezoresistive and piezoelectric measurements at the 3D force sensor. In addition, an Extended direct Piezoresistive/Piezoelectric fusion (EPPF) algorithm is introduced to enhance the DPPF method via performing the fusion in a range of frequencies instead of a particular one. The DPPF and EPPF methods were implemented on the 3D force sensor data, and the developed fusion algorithms were tested on the new 3D force sensor via experimental data. The simulation results show that the proposed fusion methods have been effective in achieving lower Root Mean Square Error (RMSE) in the estimated strain than those obtained from the tuned nlhw models at different operating frequencies.
  • Thumbnail Image
    Item
    RCN2: Residual Capsule Network V2
    (IEEE Xplore, 2021-06) Anilkumar, Arjun Narukkanchira; El-Sharkawy, Mohamed; Electrical and Computer Engineering, School of Engineering and Technology
    Unlike Convolutional Neural Network (CNN), which works on the shift-invariance in image processing, Capsule Networks can understand hierarchical model relations in depth[1]. This aspect of Capsule Networks let them stand out even when models are enormous in size and have accuracy comparable to the CNNs, which are one-tenth of its size. The capsules in various capsule-based networks were cumbersome due to their intricate algorithm. Recent developments in the field of Capsule Networks have contributed to mitigating this problem. This paper focuses on bringing one of the Capsule Network, Residual Capsule Network (RCN) to a comparable size to modern CNNs and thus restating the importance of Capsule Networks. In this paper, Residual Capsule Network V2 (RCN2) is proposed as an efficient and finer version of RCN with a size of 1.95 M parameters and an accuracy of 85.12% for the CIFAR-10 dataset.
  • Thumbnail Image
    Item
    RCNN-SliceNet: A Slice and Cluster Approach for Nuclei Centroid Detection in Three-Dimensional Fluorescence Microscopy Images
    (IEEE, 2021) Wu, Liming; Han, Shuo; Chen, Alain; Salama, Paul; Dunn, Kenneth W.; Delp, Edward J.; Electrical and Computer Engineering, School of Engineering and Technology
    Robust and accurate nuclei centroid detection is important for the understanding of biological structures in fluorescence microscopy images. Existing automated nuclei localization methods face three main challenges: (1) Most of object detection methods work only on 2D images and are difficult to extend to 3D volumes; (2) Segmentation-based models can be used on 3D volumes but it is computational expensive for large microscopy volumes and they have difficulty distinguishing different instances of objects; (3) Hand annotated ground truth is limited for 3D microscopy volumes. To address these issues, we present a scalable approach for nuclei centroid detection of 3D microscopy volumes. We describe the RCNN-SliceNet to detect 2D nuclei centroids for each slice of the volume from different directions and 3D agglomerative hierarchical clustering (AHC) is used to estimate the 3D centroids of nuclei in a volume. The model was trained with the synthetic microscopy data generated using Spatially Constrained Cycle-Consistent Adversarial Networks (SpCycle-GAN) and tested on different types of real 3D microscopy data. Extensive experimental results demonstrate that our proposed method can accurately count and detect the nuclei centroids in a 3D microscopy volume.