A generic nuclei detection method for histopathological breast images

The detection of cell nuclei plays a key role in various histopathological image analysis problems. Considering the high variability of its applications, we propose a novel generic and trainable detection approach. Adaption to specific nuclei detection tasks is done by providing training samples. A trainable deconvolution and classification algorithm is used to generate a probability map indicating the presence of a nucleus. The map is processed by an extended watershed segmentation step to identify the nuclei positions. We have tested our method on data sets with different stains and target nuclear types. We obtained F1-measures between 0.83 and 0.93.

[1]  Luca Maria Gambardella,et al.  Mitosis Detection in Breast Cancer Histology Images with Deep Neural Networks , 2013, MICCAI.

[2]  Marcel van Herk.  A fast algorithm for local minimum and maximum filters on rectangular and octagonal kernels , 1992, Pattern Recognit. Lett..

[3]  Lin Yang,et al.  Robust Segmentation of Overlapping Cells in Histopathology Specimens Using Parallel Seed Detection and Repulsive Level Set , 2012, IEEE Transactions on Biomedical Engineering.

[4]  A. Ruifrok,et al.  Quantification of histochemical staining by color deconvolution. , 2001, Analytical and quantitative cytology and histology.

[5]  Chanho Jung,et al.  Unsupervised Segmentation of Overlapped Nuclei Using Bayesian Classification , 2010, IEEE Transactions on Biomedical Engineering.

[6]  A. Huisman,et al.  Automatic Nuclei Segmentation in H&E Stained Breast Cancer Histopathology Images , 2013, PloS one.

[7]  Yousef Al-Kofahi,et al.  Improved Automatic Detection and Segmentation of Cell Nuclei in Histopathology Images , 2010, IEEE Transactions on Biomedical Engineering.

[8]  Hui Kong,et al.  Partitioning Histopathological Images: An Integrated Framework for Supervised Color-Texture Segmentation and Cell Splitting , 2011, IEEE Transactions on Medical Imaging.

[9]  Gabriela Alexe,et al.  High expression of lymphocyte-associated genes in node-negative HER2+ breast cancers correlates with lower recurrence rates. , 2007, Cancer research.

[10]  Daniel Heim,et al.  Detection and Segmentation of Cell Nuclei in Virtual Microscopy Images: A Minimum-Model Approach , 2012, Scientific Reports.

[11]  Serge Beucher,et al.  The Morphological Approach to Segmentation: The Watershed Transformation , 2018, Mathematical Morphology in Image Processing.

[12]  Anant Madabhushi,et al.  An Integrated Region-, Boundary-, Shape-Based Active Contour for Multiple Object Overlap Resolution in Histological Imagery , 2012, IEEE Transactions on Medical Imaging.

[13]  Andrew Zisserman,et al.  Learning to Detect Cells Using Non-overlapping Extremal Regions , 2012, MICCAI.

[14]  John Meyer,et al.  Grading nuclear pleomorphism on histological micrographs , 2008, 2008 19th International Conference on Pattern Recognition.

[15]  Hervé Delingette,et al.  Medical Image Computing and Computer-Assisted Intervention – MICCAI 2012 , 2012, Lecture Notes in Computer Science.

[16]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[17]  Rasmus Larsen,et al.  Learning histopathological patterns , 2011, Journal of pathology informatics.