TY - JOUR
T1 - Spatially Constrained Context-Aware Hierarchical Deep Correlation Filters for Nucleus Detection in Histology Images
AU - Javed, Sajid
AU - Mahmood, Arif
AU - Dias, Jorge
AU - Werghi, Naoufel
AU - Rajpoot, Nasir
N1 - Funding Information:
This publication acknowledges the support provided by the Khalifa University of Science and Technology under Award No. RC1-2018-KUCARS and electrical engineering and computer science departmnt at KU.The last author of this work, Nasir Rajpoot (NR), is supported by the UK Medical Research Council (No. MR/P015476/1). NR is also supported by the PathLAKE digital pathology consortium, which is funded from the Data to Early Diagnosis and Precision Medicine strand of the government’s Industrial Strategy Challenge Fund, managed and delivered by UK Research and Innovation (UKRI).
Publisher Copyright:
© 2021
PY - 2021/8
Y1 - 2021/8
N2 - 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.
AB - 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.
KW - Computational pathology
KW - Correlation filters
KW - Deep learning
KW - Nucleus detection
UR - http://www.scopus.com/inward/record.url?scp=85109140584&partnerID=8YFLogxK
U2 - 10.1016/j.media.2021.102104
DO - 10.1016/j.media.2021.102104
M3 - Article
C2 - 34242872
AN - SCOPUS:85109140584
SN - 1361-8415
VL - 72
JO - Medical Image Analysis
JF - Medical Image Analysis
M1 - 102104
ER -