A Novel Measure for Nuclear Pleomorphism in Lung Adenocarcinoma

Najah Alsubaie1,2, Shan E Ahmed Raza1, David Snead3, Nasir M. Rajpoot1
1University of Warwick 2Princess Nourah University 3University Hospitals Coventry and Warwickshire

Nuclear morphology is used as a significant cue in different histology grading systems such as grading of breast cancer [1]. At the early stages of cancer, tumour nuclei have similar shape, size and texture. At the advanced stages, tumour nuclei deform into non-uniform shapes and unequal sizes. In lung cancer, several studies have shown that nuclear pleomorphism is a potential prognostic indicator and could be correlated with patient survival [2, 3]. Pathologists normally score nuclear pleomorphism by comparing them with the normal cells. They typically observe nuclei size, shape and the visibility of nucleoli [1]. This scoring mechanism is hugely affected by the subjectivity of human perception, as enormous variations in nuclear morphology make it very difficult for human observer to describe it in a precise way. Therefore, the reproducibility and reliability of these scoring systems are uncertain [4, 5].

In this study, we provide an automatic and objective method to measure the variability of nuclei shapes in tumour regions to overcome the limitations of the current clinical routine in nuclei scoring. The proposed method includes two stages: 1) Nuclear segmentation and 2) Nuclear shape analysis. We extract a 500×500 visual field from an image of Haematoxylin & Eosin (H&E) stained slide acquired with Omnyx VL120 scanner at 40× magnifications. We segment nuclear structures [6] by extracting the Haematoxylin channel using stain deconvolution algorithm [7]. To refine our input, we perform detection and classification of the extracted structures to remove all the segmented structures that have not been detected and classified as tumour nuclei [8]. In the second stage, we preform nuclear shape analysis to model each of the segmented nuclei. For this, we first calculate the turning angles at uniformly distanced sample points on nuclear boundary. We then encode the nuclear shape as Weighted Entropy (WE) of the turning angles. The probability distribution of WE of all the nuclei in a visual field is then used to describe the nuclear shape in tumour regions, see Figure 1. The proposed approach is able to provide an automatic and objective way to analyse nuclear morphology in tumour regions. Incorporating this approach into the analysis of whole slide images of lung adenocarcinoma would improve the reliability of using nuclear morphology in assessing tumour regions.



[1]    M. Veta, et al., “Breast cancer histopathology image analysis: a review,” IEEE Trans. Biomed. Eng., pp. 1400–11, 2014.

[2]    J. A. Barletta, et al., “Prognostic significance of grading in lung adenocarcinoma,” Cancer, pp. 659–69, Feb. 2010.

[3]   J. H. von der Thüsen, et al., “Prognostic significance of predominant histologic pattern and nuclear grade in resected adenocarcinoma of the lung: potential parameters for a grading system,” J. Thorac. Oncol., pp. 37–44, 2013.

[4]    Y. Nakazato, et al., “Interobserver agreement in the nuclear grading of primary pulmonary adenocarcinoma.,” J. Thorac. Oncol., pp. 736–43, 2013.

[5]     B. Dunne, et al., “Scoring nuclear pleomorphism in breast cancer,” Histopathology, pp. 259–265, 2001.

[6]     M. Veta, et al., “Automatic Nuclei Segmentation in H&E Stained Breast Cancer Histopathology Images,” PLoS One, p. e70221, 2013.

[7]     A. M. Khan, et al., “A nonlinear mapping approach to stain normalization in digital histopathology images using image-specific color deconvolution.,” IEEE Trans. Biomed. Eng., pp. 1729–38, 2014.

[8]   K. Sirinukunwattana, et al., “A Spatially Constrained Deep Learning Framework for Detection of Epithelial Tumor Nuclei in Cancer Histology Images,” Springer International Publishing, , pp. 154–162, 2015.








Figure 1. First column shows the original images extracted from a high grade tumour (top) and low grade tumour (bottom). The second column shows the segmented images. Histograms in third column show the marginal distribution of WE, which represents the morphology of nuclei in each image. We can notice that the probability distribution of the high-grade tumour (top) has a higher median value (1.376) of the computed WE with a wider range of margin (0.4 to 2.4). The low-grade tumour (bottom) region gives lower median value (1.231) of the WE and a relatively smaller margin (0.7-2). Thus, well differentiated tumour regions will have lower median value and smaller margin of the WE compared to the poorly differentiated.




An Integrated Environment for Morphometric Profiling of Tumor

Talha Qaiser1, Korsuk Siriniukunwattana1, Nasir M. Rajpoot1
1University of Warwick

The adoption of digital imaging in histopathology is making multi-gigapixel whole-slide images (WSIs) available for conducting tissue morphometric analyses. In routine, the experts visually examine the stained tissue slides under the conventional microscope and this method has been exercised for several decades with high inter-observer and even intra-observer variability, rendering the diagnosis often non-reproducible [1-3]. An integrated framework to extract quantitative morphological features from histology images and perform morphometric analyses that can lead to the identification of outcome-related features and provide a more accurate and reproducible means to assess cancer. In this talk, we will describe the architecture of the proposed environment and discuss various modules that are being incorporated into the environment.


The proposed integrated environment enables the analytics of whole-slide tissue profiling for multiple cancer cases. It can also facilitate the studies of prognostic models, in which morphometric features strongly correlated with the outcome of cancers can be identified and used as image-based markers. This integrated framework is mainly based on three main components: (1) core module comprising of visualization of WSIs at multiple magnification levels, enabling display of clinical and imaging data for multiple cases simultaneously, (2) analytical module contains an interactive tool for measuring dimensions of tissue components and interactive annotation module, and (3) analytics module applied to data from a selected subset of cases or all the cases using quantitative morphological measurements including those derived from automatic phenotyping of cells [4]. The proposed environment can be further extended by adding new analytical modules and it can directly bring the benefits of quantitative analysis into pathological practices, thereby increasing reproducibility of cancer diagnosis.



Smits, Alexander JJ, et al. "The estimation of tumor cell percentage for molecular testing by pathologists is not accurate." Modern Pathology 27(2):168-174 (2014).

Viray, Hollis, et al. "A prospective, multi-institutional diagnostic trial to determine pathologist accuracy in estimation of percentage of malignant cells." Archives of Pathology & Laboratory Medicine 137(11):1545-1549 (2013).

Winters, Bradford, et al. "Diagnostic errors in the intensive care unit: a systematic review of autopsy studies." BMJ Quality & Safety 21(11):894-902 (2012).

Sirinukunwattana, Korsuk, et al. "Locality Sensitive Deep Learning for Detection and Classification of Nuclei in Routine Colon Cancer Histology Images." IEEE Transactions on Medical Imaging special issue on Deep Learning in Medical Imaging 35(5):1196-1206 (2016)

Comparison of vascular networks from high resolution 3D whole organ microscopic analysis

Mike Pesavento1, Pranathi V. N. Vemuri1, Caroline Miller1, Jenny Folkesson1, Megan Klimen1

Understanding hemodynamics in circulatory systems is a critical component to identifying pathophysiologic states in tissue. Significant progress has been made in vascular network imaging; resolution has increased for high volume methods (eg microCT and MRI), and volume has increased for high resolution methods (eg multi-photon and confocal microscopy). 3Scan’s Knife Edge Scanning Microscope (KESM) spans the gap between high volume and high resolution imaging modalities[1].


Bright field images of resin-embedded, whole-organs (brain and pancreas) were obtained from mice following systemic perfusion with India Ink. Images are taken with a resolution of 0.7 um per pixel in XY and a typical slice depth of 5 um in Z, enabling large-scale analysis and comparison of vascular networks of whole organs consisting of up to 5 TB of imaging data in 3D and a maximum physical volume of 50 x 50 x 20 mm. Vascular features are identified via parallelized vessel segmentation and vectorization methods.


Comparison of vascular features within a single organ reveals significant differences between the area analyzed within target tissue, largely as a result of the fractal dimension of the vascular network. Comparison of vascular network features between organs yields significant differences between vascular networks that are commensurate with the function of the vascular network for that organ.


Rapid throughput analysis of high volume vascular data provides an unprecedented ability to  compare vascular features between different vascular networks, as well as identify pathological states within those networks.

[1] Pesavento MJ, Miller C, Pelton K, Maloof M, Monteith CE, Vemuri VNP, Klimen M (submitted). “Knife-Edge Scanning Microscopy for Multi-cubic Centimeter Microscopic Analysis of Microvasculature”, Microscopy Today

Cytomine for collaborative and semantic analysis of digital pathology images

Raphaël Marée1, Loic Rollus1, Renaud Hoyoux1, Benjamin Stévens1, Gilles Louppe1, Rémy Vandaele1, Jean-Michel Begon1, Philipp Kainz2, Pierre Geurts1, Louis Wehenkel1
1University of Liège 2Medical University of Graz

Cytomine (http://www.cytomine.be/) is an open-source, rich internet application, for remote visualization of whole-slide images (à la Google Maps), collaborative and semantic annotation of regions of interest using user-defined ontologies, and semi-automated image analysis using machine learning.


Here we will describe our design choices that allow data scientists and image analysis software developers to use and extend the
platform in various ways. In particular we will describe our vocabulary-driven annotations of images, HTTP based RESTful API to
import/export data through web services, and our supervised learning workflows including our semantic proofreading tools for object classification, image segmentation, content-based image retrieval, and landmark detection.


We will then brielfy present our latest applications of the software as it is now being actively used by many research groups working on large sets of images in lung/breast cancer research, renal pathology, toxicology and developmental studies, ... (see publication list: http://www.cytomine.be/#publications).

Learning based detection of early neoplastic changes in histological images

Mira Valkonen1,2, Matti Nykter1, Leena Latonen1, Pekka Ruusuvuori3
1BioMediTech, University of Tampere 2Tampere University of Technology, Finland 3Tampere University of Technology, Pori

Digital pathology has been rapidly expanding into a routine practice, which has enabled the development of image analysis tools for quantification of histological images. Prostatic intraepithelial neoplasia (PIN) represents premalignant tissue involving epithelial growth confined in the lumen of prostatic acini. In the attempts to understand oncogenesis in the human prostate, we studied early neoplastic changes in mouse prostatic intraepithelial neoplasia (mPIN) confined in prostate. We implemented an image analysis pipeline for describing early morphological changes in hematoxylin and eosin (H&E) stained histological images. The model is based on manually engineered features and supervised learning with random forest model. For training, we used a set of mPIN lesions of abnormal epithelial cell growth and glands of normal tissue segmented by an expert. The extracted features include 102 local descriptors related to tissue texture and spatial arrangement and distribution of nuclei. These extracted features provide a numerical representation of a tissue sample and were used to computationally learn a discriminative model using machine learning. The implemented random forest model is an ensemble of 50 classification trees and it uses bootstrap aggregation to improve stability and accuracy. Leave-one-out cross-validation (LOOCV) was used to evaluate the performance of our random forest model. The classification model was able to discriminate normal tissue segments from early mPIN lesions and also describe the spatial heterogeneity of the tissue samples. The model can be easily interpreted and used to assess the contribution of individual features. This feature significance provides information about differences in the histology between normal glands and early neoplastic lesions.

Nobody likes the chubby peewee

Jennifer Scheidel1, Hendrik Schäfer1, Jörg Ackermann1, Marie Hebel1, Tim Schäfer1, Claudia Döring2, Sylvia Hartmann2, Martin-Leo Hansmann2, Ina Koch1
1Johann Wolfgang Goethe-University Frankfurt am Main 2Goethe-University Hospital

We present an analysis of the spatial distribution of Hodgkin and Reed-Sternberg cells in classical Hodgkin lymphoma. Hodgkin lymphoma is a tumor of the lymphatic system. Large tumor cells called Hodgkin/Reed-Sternberg (HRS) cells characterize the classical Hodgkin lymphoma (cHL). Typically, in round numbers only 1 % of the lymph node are HRS cells. Clinical diagnosis generates a large number of histological images in which HRS cells are immunohistochemically stained by CD30. Such images are snap shots of the disease available for a broad variety of medical cases and offer the opportunity to systematically study the morphology and spatial distribution of HRS cells in the tissue.  The automated analysis of images of histological tissues may enable for valuable conclusions on the co-operative migration behavior of malignant cells within a lymph node. We analyzed in total 35 images of tissue sections of the cHL subtypes, nodular sclerosis (NScHL) and mixed cellularity (MCcHL) as well as images of an inflammation of the lymph node called lymphadenitis (LA) [1]. Our imaging pipeline identified the profiles of in round numbers 400.000 CD30 positive cells in the tissue sections [2]. The distribution of the diameter of the cells had its maximum in the range of 20 to 22.5 μm for cHL and of 15-17.5 μm for LA. The estimated mean diameter of HRS cell profiles in NScHL was 30.6±10.2 μm, whereas the mean diameter for MCcHL was slightly smaller, i.e., 28.6±9.3 μm. Further, we assigned each individual cell to one of eight predefined classes according to the morphological features, eccentricity, solidity, and area of its profile and analyzed the neighbor relations of the cells belonging to different profile classes.  In the choice of their next neighbors, the cells had statistical significant preferences and aversions for distinct classes. Each class of cell exhibited specific patterns of preferences and aversions, e.g., round and small cells tended to stay in the neighborhood of its own kind and were shunned by cells of other classes. The patterns of preferences and aversions were differently pronounced depending on the medical diagnosis. We analyzed the distribution of distances to the nearest neighbor to check whether attractive or repulsive forces between cells of specific classes were the source of the patterns of preferences and aversions. The distribution of distances proved a clustering of the cells in the tissue but, e.g., the comparable large mean distance between small and round cells contradicted the hypotheses of an attraction that forces small and round cell profiles to stay among their own kind. The influence of the complex structure of the lymph node and specific cell interactions, e.g., by chemokines and cytokines, are possible explanations for the overall clustering of the cell profiles. The patterns of preferences and aversions of the specific profiles class were more likely an effect of different motilities of the cells in the tissue.


[1] Schäfer, T., Schäfer, H., Schmitz, A., Ackermann, J., Dichter, N., Döring, C., Hartmann, S., Hansmann, M.L., and Koch, I. (2013). Image database analysis of Hodgkin lymphoma. Computational Biology and Chemistry, 46, 1-7.

[2] Schäfer, H., Schäfer, T.,  Ackermann, J., Dichter, N., Döring, C., Hartmann, S., Hansmann, M.L., and Koch, I. (2015). CD30 cell graphs of Hodgkin lymphoma are not scale-free—an image analysis approach. Bioinformatics, 32, 122-129.

Recognizing the BRAF mutant-like tumors from whole-slide pathology images

Vlad Popovici1
1Masaryk University, Brno


Tumor heterogeneity plays a central role in the observed variability of treatment responses and survival of cancer patients. At the same time, it represents a major hurdle on the path towards a personalized medicine, with a plethora of molecular biomarkers being recently proposed to partially resolve this heterogeneity. This is the case for the BRAF mutant-like (BL) gene expression signature [1], which identifies a high risk subpopulation of colorectal cancers (CRCs). These tumors, while not harboring the BRAF V600E mutation, display a similar pattern of gene activation (for a selected set of genes) with the mutants and, more importantly, share the same dismal outcome. It is thus of great importance that the BL tumors are identified early on and currently the only method relies on a 64-gene signature [1], which is not yet implemented in clinical practice. We propose to build a tissue-based proxy biomarker which would provide an indication whether molecular testing should be performed and which could be integrated in the daily practice without disturbing any protocol, since it would rely on routine H&E-stained slides. We will restrict, for the moment, this tissue biomarker to stage III, microsatellite-stable (MSS) CRCs, which form a more homogeneous subpopulation.


Methods and Results

The data collection consisted of n=113 samples for which both histopathology whole-slide images and clinical data were available, along with the corresponding BL status (a real-valued score, with positive values indicating BRAF mutant-like cases). All samples are stage III, MSS CRCs. The collection was divided into a training (n=40) and testing (n=73) disjoint sets. The images were scanned at 40x magnification and later down-scaled to an equivalent of 2.5x. Tumor regions were extracted based on expert annotation and the color images were further denoised (Gaussian filtering) and hematoxylin intensity estimated via color deconvolution [2]. All later processing was performed only using these gray scale (hematoxylin intensity) images.

Local feature descriptors (vectors of d=64 values) were generated using the  SURF method [3] and a bag of features [4] representation generated for each image, based on a dictionary of size k=50 (optimized on the training set). A DLDA (diagonal LDA) classifier was trained to predict the BL status (binary classification).

The dictionary consisted in k=50 image feature vectors corresponding to patches of sizes varying between 14x14 and 54x54, from highly variable (high content) regions of the images. Of these, 9 feature vectors were significantly associated with BL status (t-test and correlation test p < 0.05) and also with the mucinous status of the tumors. The DLDA classifier was built on 30 variables (image features, the number was optimized via cross-validation on the training set) and achieved an accuracy on the test set of 91.78% (95% CI=82.89-96.49) corresponding to a sensitivity of 93.75% and a specificity of 90.25% (6 misclassified samples out of 73). The stratification induced by the classifier was marginally significant in survival analysis (survival after relapse): HR=1.62, p=0.06. For the same set of patients, the molecular biomarker has HR=2.22, p=0.02.



On a relatively small data set we were able to build an image-based proxy biomarker for BL CRCs achieving good test performance. This biomarker may provide a starting point for a screening test (e.g. by adjusting its threshold) for identifying additional high risk patients. Since it uses standard histopathology images and by integration with other automatic image analysis tools (e.g. tumor region identification), the proposed method can be integrated in daily clinical practice without disturbing the protocols in place and can work autonomously to provide complementary diagnostic and prognostic information.



[1] V. Popovici et al. Identification of a poor-prognosis BRAF-mutant-like population of patients with colon cancer. J Clin Oncol 2012, 30(12):1288-1295

[2] AC Ruifrok, DA Johnston. Quantification of histochemical staining by color deconvolution. Anal Quant Cytol. 2001, 23(4):291-299  

[3] H. Bay et al. SURF: speeded up robust features. Comput Vis Image Und. 2008,  110(3): 346-359. 

[4]  G. Csurka G et al. Visual categorization with bags of keypoints. ECCV International Workshop on Statistical Learning in Computer Vision, 2004.

Shedding a Different Light on Disease: An Introduction to Infrared Based Spectral Pathology

Alex Henderson1, Peter Gardner1
1University of Manchester

Given that the eye is an excellent photon detector and the human brain is one of the most advanced image processing systems known to man it is not surprising that visible light microscopy has been the mainstay of pathological analysis. However significant advances in detector technology and computer processing power make other regions of the electromagnetic spectrum attractive for tissue analysis. In this presentation I will introduce the revolutionary new techniques of infrared based technology that can facilitate detailed tissue analysis. Hyperspectral imaging coupled with sophisticated computer algorithms enable cancerous tissue to be indentified and graded and, in favourable cases, an indication of prognosis to be obtained. This technique lends itself to automation and would be particularly useful for screening large numbers of biopsy samples for the common types of cancer.

Whole slide scanning to speed up and improve pancreatic beta cell volume quantification

Willem Staels1, Gunter Leuckx1, Yves Sucaet1, Yves Heremans1, Nico De Leu1, Peter In't Veld1, Harry Heimberg1
1Vrije Universiteit Brussel

Quantifying the beta cell volume is an essential tool for the study of diabetes. For this, researchers currently rely on laborious immunohistochemical and bioinformatical analyses of pancreatic tissues. Most time-consuming is the image acquisition step from stained paraffin embedded tissue slides to digital images of entire tissue sections. Whole slide scanning is revolutionizing clinical pathology and research as it allows for fast and effortless acquisition of high resolution images. We compared current methods relying on inverted microscopy with whole slide scanning technology in combination with the Pathomation software platform for quantification of the pancreatic beta cell volume. In both nonpregnant and pregnant mice beta cell volumes measured via both methods are highly similar, while the scanning technique proves to be up to 4 times faster. Whole slide scanning offers a reliable and faster alternative for quantification of the pancreatic beta cell volume.