Spatially Constrained Context-Aware Hierarchical Deep Correlation Filters for Nucleus Detection in Histology Images |
| |
| Institution: | 1. Khalifa University Center for Autonomous Robotic Systems (KUCARS), Khalifa University, Abu Dhabi, UAE.;2. Department of Computer Science, Information Technology University, Lahore, Pakistan.;3. Department of Computer Science, University of Warwick, Coventry, CV4 7AL, U.K.;4. Department of Pathology, University Hospitals Coventry and Warwickshire, Walsgrave, Coventry, CV2 2DX, U.K.;5. The Alan Turing Institute, London, NW1 2DB, U.K.;1. Applied Tumor Immunity Clinical Cooperation Unit, National Center for Tumor Diseases, German Cancer Research Center, Heidelberg, Germany;2. Statistical Physics and Theoretical Biophysics Group, Institute for Theoretical Physics, Heidelberg University, Heidelberg, Germany;3. Department of Medical Oncology, National Center for Tumor Diseases, University Hospital Heidelberg, Heidelberg, Germany;4. Department of Medicine II, Faculty of Medicine at Mannheim, Heidelberg University, Mannheim, Germany;5. Department of Informatics, Technical University Federico Santa María, Valparaiso, Chile;1. Department of Computer Science, Stony Brook University, Stony Brook, NY, USA;2. Montreal Institute for Learning Algorithms, University of Montreal, Montreal, Canada;3. Department of Biomedical Informatics, Stony Brook University, Stony Brook, NY, USA;4. Oak Ridge National Laboratory, Oak Ridge, TN, USA;5. Department of Pathology, Stony Brook University Medical Center, Stony Brook, NY, USA;6. School of Biomedical Engineering, Health Science Center, Shenzhen University, China;7. Center for Biomedical Imaging & Informatics, Rutgers, the State University of New Jersey, New Brunswick, NJ, USA;8. Rutgers Cancer Institute of New Jersey, Rutgers, the State University of New Jersey, NJ, USA;9. Division of Medical Informatics, Rutgers-Robert Wood Johnson Medical School, Piscataway Township, NJ, USA;10. Cancer Center, Stony Brook University Hospital, Stony Brook, NY, USA;1. IBM Zurich Research Lab, Zurich, Switzerland;2. Computer-Assisted Applications in Medicine, ETH Zurich, Zurich, Switzerland;1. Pattern Recognition Lab, DCIS, PIEAS, Nilore, Islamabad 45650, Pakistan;2. PIEAS Artificial Intelligence Center (PAIC), PIEAS, Nilore, Islamabad 45650, Pakistan;3. Deep Learning Lab, Centre for Mathematical Sciences, PIEAS, Nilore, Islamabad 45650, Pakistan;4. Department of Electronic Engineering, Faculty of Engineering and Green Technology, Universiti Tunku Abdul Rahman, Kampar 31900, Malaysia;5. International Islamic University, Islamabad, Pakistan |
| |
| Abstract: | Nucleus detection in histology images is a fundamental step for cellular-level analysis in computational pathology. In clinical practice, quantitative nuclear morphology can be used for diagnostic decision making, prognostic stratification, and treatment outcome prediction. Nucleus detection is a challenging task because of large variations in the shape of different types of nucleus such as nuclear clutter, heterogeneous chromatin distribution, and irregular and fuzzy boundaries. To address these challenges, we aim to accurately detect nuclei using spatially constrained context-aware correlation filters using hierarchical deep features extracted from multiple layers of a pre-trained network. During training, we extract contextual patches around each nucleus which are used as negative examples while the actual nucleus patch is used as a positive example. In order to spatially constrain the correlation filters, we propose to construct a spatial structural graph across different nucleus components encoding pairwise similarities. The correlation filters are constrained to act as eigenvectors of the Laplacian of the spatial graphs enforcing these to capture the nucleus structure. A novel objective function is proposed by embedding graph-based structural information as well as the contextual information within the discriminative correlation filter framework. The learned filters are constrained to be orthogonal to both the contextual patches and the spatial graph-Laplacian basis to improve the localization and discriminative performance. The proposed objective function trains a hierarchy of correlation filters on different deep feature layers to capture the heterogeneity in nuclear shape and texture. The proposed algorithm is evaluated on three publicly available datasets and compared with 15 current state-of-the-art methods demonstrating competitive performance in terms of accuracy, speed, and generalization. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
