Clustering methods applied in the detection of Ki67 hot‐spots in whole tumor slide images: An efficient way to characterize heterogeneous tissue‐based biomarkers

Whole‐slide scanners allow the digitization of an entire histological slide at very high resolution. This new acquisition technique opens a wide range of possibilities for addressing challenging image analysis problems, including the identification of tissue‐based biomarkers. In this study, we use whole‐slide scanner technology for imaging the proliferating activity patterns in tumor slides based on Ki67 immunohistochemistry. Faced with large images, pathologists require tools that can help them identify tumor regions that exhibit high proliferating activity, called “hot‐spots” (HSs). Pathologists need tools that can quantitatively characterize these HS patterns. To respond to this clinical need, the present study investigates various clustering methods with the aim of identifying Ki67 HSs in whole tumor slide images. This task requires a method capable of identifying an unknown number of clusters, which may be highly variable in terms of shape, size, and density. We developed a hybrid clustering method, referred to as Seedlink. Compared to manual HS selections by three pathologists, we show that Seedlink provides an efficient way of detecting Ki67 HSs and improves the agreement among pathologists when identifying HSs. © 2012 International Society for Advancement of Cytometry

[1]  Alberto Maria Segre,et al.  Programs for Machine Learning , 1994 .

[2]  Anil K. Jain.  Data Clustering: User's Dilemma , 2007, MLDM.

[3]  Sabrina Oliveira,et al.  Recent advances in molecular imaging biomarkers in cancer: application of bench to bedside technologies. , 2010, Drug discovery today.

[4]  C. Scopa,et al.  Expression of cell cycle inhibitors p21, p27, p14 and p16 in gliomas. Correlation with classic prognostic factors and patients' outcome , 2008, Neuropathology : official journal of the Japanese Society of Neuropathology.

[5]  A. Berger.  FUNDAMENTALS OF BIOSTATISTICS , 1969 .

[6]  B. Gonçalves,et al.  Expression of Ki‐67, Topoisomerase IIα and c‐MYC in astrocytic tumors: Correlation with the histopathological grade and proliferative status , 2006, Neuropathology : official journal of the Japanese Society of Neuropathology.

[7]  Daniel A. Keim,et al.  An Efficient Approach to Clustering in Large Multimedia Databases with Noise , 1998, KDD.

[8]  P Kronqvist,et al.  Ki67 immunohistochemistry: a valuable marker in prognostication but with a risk of misclassification: proliferation subgroups formed based on Ki67 immunoreactivity and standardized mitotic index , 2006, Histopathology.

[9]  A. Madabhushi,et al.  Histopathological Image Analysis: A Review , 2009, IEEE Reviews in Biomedical Engineering.

[10]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[11]  Gerald C. Chu,et al.  P53 and Pten control neural and glioma stem/progenitor cell renewal and differentiation , 2008, Nature.

[12]  Stanley Wasserman,et al.  Social Network Analysis: Methods and Applications , 1994, Structural analysis in the social sciences.

[13]  Philip Chan,et al.  Determining the number of clusters/segments in hierarchical clustering/segmentation algorithms , 2004, 16th IEEE International Conference on Tools with Artificial Intelligence.

[14]  M. A. Wong,et al.  A Hybrid Clustering Method for Identifying High-Density Clusters , 1982 .

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

[16]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.

[17]  S Wolfsberger,et al.  Ki67 index in intracranial ependymoma: a promising histopathological candidate biomarker , 2008, Histopathology.

[18]  Benoît Plancoulaine,et al.  A simple way of quantifying immunostained cell nuclei on the whole histologic section , 2003, Cytometry. Part A : the journal of the International Society for Analytical Cytology.

[19]  Jun Kong,et al.  Computer-aided prognosis of neuroblastoma on whole-slide images: Classification of stromal development , 2009, Pattern Recognit..

[20]  Joachim M. Buhmann,et al.  Computational Pathology: Challenges and Promises for Tissue Analysis , 2015, Comput. Medical Imaging Graph..

[21]  Isabelle Salmon,et al.  KI-67 hot-spots detection on glioblastoma tissue sections , 2010, 2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[22]  Kevin Françoisse,et al.  The Sum-over-Paths Covariance Kernel: A Novel Covariance Measure between Nodes of a Directed Graph , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[23]  Isabelle Salmon,et al.  Requirements for the valid quantification of immunostains on tissue microarray materials using image analysis , 2009, Proteomics.

[24]  Martin G. Everett,et al.  A Graph-theoretic perspective on centrality , 2006, Soc. Networks.

[25]  Christine Haberler,et al.  Histopathologic assessment of hot‐spot microvessel density and vascular patterns in glioblastoma: Poor observer agreement limits clinical utility as prognostic factors , 2006, Cancer.

[26]  Leonard M. Freeman,et al.  A set of measures of centrality based upon betweenness , 1977 .

[27]  E. Englund,et al.  Different assessments of immunohistochemically stained Ki-67 and hTERT in glioblastoma multiforme yield variable results: a study with reference to survival prognosis. , 2008, Clinical neuropathology.

[28]  Mark E. J. Newman.  A measure of betweenness centrality based on random walks , 2005, Soc. Networks.

[29]  Nicolas Loménie,et al.  Time-efficient sparse analysis of histopathological whole slide images , 2011, Comput. Medical Imaging Graph..