Skip to content
Publicly Available Published by De Gruyter December 20, 2018

Histo-genomics: digital pathology at the forefront of precision medicine

  • Ivraym Barsoum , Eriny Tawedrous , Hala Faragalla and George M. Yousef EMAIL logo
From the journal Diagnosis

Abstract

The toughest challenge OMICs face is that they provide extremely high molecular resolution but poor spatial information. Understanding the cellular/histological context of the overwhelming genetic data is critical for a full understanding of the clinical behavior of a malignant tumor. Digital pathology can add an extra layer of information to help visualize in a spatial and microenvironmental context the molecular information of cancer. Thus, histo-genomics provide a unique chance for data integration. In the era of a precision medicine, a four-dimensional (4D) (temporal/spatial) analysis of cancer aided by digital pathology can be a critical step to understand the evolution/progression of different cancers and consequently tailor individual treatment plans. For instance, the integration of molecular biomarkers expression into a three-dimensional (3D) image of a digitally scanned tumor can offer a better understanding of its subtype, behavior, host immune response and prognosis. Using advanced digital image analysis, a larger spectrum of parameters can be analyzed as potential predictors of clinical behavior. Correlation between morphological features and host immune response can be also performed with therapeutic implications. Radio-histomics, or the interface of radiological images and histology is another emerging exciting field which encompasses the integration of radiological imaging with digital pathological images, genomics, and clinical data to portray a more holistic approach to understating and treating disease. These advances in digital slide scanning are not without technical challenges, which will be addressed carefully in this review with quick peek at its future.

Introduction

Malignant neoplasms are known for their heterogeneity and evolutionary progression. The oncogenesis model described by Nowell in 1976 [1] demonstrated that when tumors accumulate mutations over time they advance to more aggressive cancers and potentially metastasize. Cancer evolution was explained further by Hanahan and Weinberg [2] who outlined the principal biological features of carcinogenesis that involve altered metabolic properties, higher mutational rates, and the capability to invade tissues and metastasize. This evolutionary model of carcinogenesis can efficiently explain the heterogeneity and multi-clonality of many cancers [3]. Different microenvironmental cues including host immune response, hypoxia, drugs and nutrient diffusion rate result in selective pressures for the development of cancer over time. To further elaborate, in the background of cancer cell high mutational rate, different heterogeneous cancer clones accumulate more mutations, become more aggressive, evade host defense systems and metastasize [4], [5], [6], [7], [8]. These features of evolution and heterogeneity set cancers as ever-moving targets, thus presenting diagnostic, prognostic and even therapeutic challenges [9]. Also, chemo- and immune-resistance, due to microenvironmental natural selection and the presence of other intrinsic factors like cancer stem cells [10], represent significant obstacles to effective cancer therapeutics [11].

Similar to cancer cell evolution, our approaches to face the cancer challenge also evolved over time. In the 1970s until early 2000s, researchers and clinician scientists focused on one gene or one biological signaling pathway at a time. However, after the era of human genome project [12], we are currently moving fast into the age of molecular profiling of cancer genomics [13], bioinformatics [14], precision medicine [15] and digital imaging that generate a wealth of “raw data” through high-throughput experimental and analytical platforms. These new advances in research tools though needed, are very complex and expanding, which sometimes makes them difficult to interpret, envision or properly utilize [16], [17]. The hope for better understanding and clinical utilization of these new tools comes through the integration of different data types [18], [19]. Pathologists, in particular, are the best fit for the job of data integration and understanding due to their unique position as a link between the bench and bedside [12], [13], [14]. This is evidenced in their emerging role in analyzing resected tumors for ancillary therapeutic predictive markers such as PD-1 [20], [21], [22].

The emergent field of digital pathology enabled the pathologists to actively contribute to better understanding of cancer pathogenesis through histo-genomics; which is simply the interface between morphology and genomics. Digital pathology can add an extra layer of information that may help us visualize, interrelate and understand the wealth of genomic, transcriptomic and proteomic data that lack spatial information [23]. Thus, digital pathology can shed more light on cancer clinical behavior, immune response and genomic data. The ultimate goal is a better quality of life for cancer patients through dissecting cancer biology, and eventually providing better therapy, which requires visualizing the evolution/progression of a malignant tumor in 4D (temporal/spatial) via histological/genomic/bioinformatics data integration.

The potential clinical utility of advanced image analysis

Histologic evaluation of tissues is the basis of anatomical pathology practice yet is famous for its inter- and intra-observer variations [24]. The subjective evaluation of histologic specimens, as currently practiced, cannot possibly capture the detailed wealth of information hidden in the heterogeneous lesional tissue in many diseases, including cancer [25]. These limitations triggered the concept of objectively quantifying morphological properties in histopathological images to aid clinical decision-making [26]. This required high-quality digital images. Older versions of slide scanning techniques demonstrated sluggish image acquisition, expensive data storing and slower networks, which in turn hindered file transfer and remote viewing. Further advancements in computer hardware and software have made slide scanning, acquiring whole slide images, digitizing of pictures and their online transfer faster and more efficient. In addition, improvements in image analysis algorithms enabled large whole slide image archives to be accessible for the extraction of quantitative and objective imaging features [26], [27], [28]. Currently, digital pathology can provide a targeted analysis of different cell components in tumor specimens and assist in tumor classification [26], [28]. In addition, image analysis algorithms can accurately define different features from the morphology of the nucleus to the characterization of composite tissues by applying different zonation and determining tissue interfaces [29], [30], [31], [32], [33], [34], [35], [36], [37], [38].

A larger number of parameters can be analyzed in a scanned digital slide/image. A partial list is shown in Table 1. An important initial step in image analysis is the segmentation of different areas of tissue interfaces [39], [40]. Segmentation involves software algorithms to differentiate between higher-order structures like interfaces between stroma/epithelium, tumor/normal tissue or areas of microvascular invasion. These always require complex techniques to describe textures or patterns in statistical terms. Currently, applying these algorithms needs sophisticated computer software hence it is hard to apply to large databases of scanned images. Other types of segmentation are less demanding, such as segmentation/delineation of cell nuclei. Different techniques are used such as color, texture thresholds, the nuclear shape and other variations [29], [37]. For example, nuclear morphometry studies can efficiently depict parameters including the size of the nucleus, its circularity, nuclear hyperchromasia, and the extent of chromatin clustering [41]. Further analysis of defined features can be used for clustering and classification of tumors by machine-learning algorithms. A representative model for the image analysis procedure is demonstrated in Figure 1 [42].

Table 1:

Examples of parameters that can be analyzed in scanned slides.

Architectural parameters:
 – Stroma/epithelium interface
 – Tumor/non-tumor interface
 – Immune cell infiltrate
 – Necrosis: localization and percent
 – Vascular distribution and microvascular proliferation
Cytological Features:
 – Nuclear size
 – Nuclear color
 – Nuclear texture
 – Nuclear contour
 – Nuclear hyperchromaticity
 – Nuclear membrane circularity
 – Extent of nuclear chromatin clumping
 – Nucleolar prominence
 – Nuclear/cytoplasmic ratio
 – Nuclear membrane irregularity
 – Tumor vascularity
Stromal features:
 – Stromal Cellularity
 – Interstitial immune response
 – Analysis of specific stromal marker expression by immunohistochemistry
Figure 1: A representative protocol for image analysis of scanned immunohistochemistry slides.(A) Original scanned image (hematoxylin and DAB). (B) Image after background correction and segmentation of cell area (green outline). Separation of hematoxylin (C) and DAB stains (D) using color deconvolution. (E) Separation of stroma (green) from epithelium (purple) using texture analysis. Reused from [42] following the guidelines of the publishing journal.
Figure 1:

A representative protocol for image analysis of scanned immunohistochemistry slides.

(A) Original scanned image (hematoxylin and DAB). (B) Image after background correction and segmentation of cell area (green outline). Separation of hematoxylin (C) and DAB stains (D) using color deconvolution. (E) Separation of stroma (green) from epithelium (purple) using texture analysis. Reused from [42] following the guidelines of the publishing journal.

To decide which character/parameter is important for diagnostic, prognostic or therapeutic purposes, two main approaches are generally used. One is the supervised or model-based approach and the other is the unsupervised or model-free approach. The model-based approaches incorporate prior knowledge of pathologists into the analyses. They employ quantitative measures of pathologically-identified morphologic models to investigate their molecular/clinical correlates. The model-free approaches do not require previously known features. Instead, they use clustering analyses to detect different histological structures and their association with cancer behavior that may reveal new features never been known to human interpretations. Further illustration of the potential use of these approaches is discussed in the following sections.

Correlation between morphological features and clinical behavior

The morphologic features such as shape, spatial arrangement, and texture of cell nuclei are known to help in determining the histological classification and grading subtypes of different tumors (as in the case of papillary thyroid cancer [43]). These features can be surrogate indices of tumor behavior or prognosis. A classic example is a prognostic value of nucleolar grading in renal cell carcinoma [44]. Advanced digital imaging was able to take this to a new level. A good example of this is central nervous system (CNS) primary tumors [45], [46], [47], [48], [49], [50]. For instance, a model-free clustering of nuclear features in digitalized slides of different glioblastomas (GBMs) revealed significant morphological correlations to patient survival. GBMs with elongated nuclei were associated with a shorter patient overall survival [25].

Another interesting example was found in breast cancer, where features that describe the morphology of the interface between the tumor and surrounding stroma was able to predict overall survival, and this was independent of other clinicopathologic and molecular factors in two independent cohorts [39]. In another study, image analysis of digital slides from triple-negative breast cancers specimens provided a prognostic model [51]. Digital pathology applications also made remarkable progress in finding grade differentiating features such as mitotic counts in prostate cancer [52], [53], [54], [55] and identifying indolent versus aggressive p16-positive oropharyngeal squamous cell carcinomas [56].

Digital image analysis was also used to correlate marker immunostaining with clinical behavior. Recent studies assessed the prognostic significance of human tissue kallikrein-related peptidases expression via image analysis of immunohistochemically stained digitalized slides in gastric cancer [42], [57]. A list of potential applications of digital pathology is shown in Table 2.

Table 2:

A list of potential applications for digital pathology analysis.

Potential applicationRepresentative studies
Predicting clinical behavior:
 – GlioblastomaCooper et al. [25]
 – Breast cancerBeck et al. [39]
 – Prostate cancerDoyle et al. [52]
 – Oropharyngeal cancerLewis et al. [56]
 – Gastric cancerKolin et al. [42]
Immune response:
 – Breast cancerYuan [58]
 – Gastric cancerFeichtenbeiner et al. [59]
 – Colorectal cancerGalon et al. [60]
Genetically subtyping cancers
 – GlioblastomaChang et al. [61]
 – Breast cancerYuan et al. [62]
3D reconstruction of solid tumors for margin assessment
 – Breast cancerClarke et al. [63]
Radio-histomics
 – Prostate cancerLanger et al. [64]

Correlation between morphological features and host immune response

The role played by tumor microenvironment during cancer evolution and the development of treatment resistance has long been acknowledged and thoroughly reviewed [56], [65], [66]. Digital pathology has been used successfully for the assessment of various microenvironmental cues, including the presence of infiltrating immune cells [67], [68], cancer-associated fibroblasts [69], [70] and stromal vasculature [71], [72], [73]. Particularly, the location of immune cells within a tumor microenvironment was found to be critical for progression in many cancers such as colorectal cancer [60], [74], [75]. Also, a number of gene expression profiles revealed marked molecular heterogeneity within the immune cell infiltrate [76], [77], [78], [79]. In breast cancer, different subclasses of lymphocytes can confer different pro- or anti-tumor functions [74], [78], [80].

Attempts to analyze immunohistochemically labeled lymphocytes in scanned/digitalized sections were performed [62], [81], [82]. In scanned digital slides from breast cancer patients, spatial statistics investigate were used to study different clustering patterns of B- and T-cells in tumor-draining and healthy lymph nodes [83]. The results revealed a higher lymphocyte clustering in lymph nodes draining tumors compared to healthy lymph nodes, with more clustering of T-cell more than B-cells, suggesting a differential role in host immune response. Digital pathology was also used to identify and quantify functionally active T-cells and natural killer cells in breast cancer [84]; and to identify the role of regulatory T cells (FOXP3+CD3+) in follicular lymphoma which was, contrary to conventional wisdom, associated with good outcome [85]. Using image analysis techniques, a functional distance range of 30–110μm between the cytotoxic and regulatory T cells showed a correlation with significant improvement in 10-year survival in 50 patients who have been diagnosed with gastric cancer [59].

Moreover, advances were made to measure distances between infiltrating lymphocytes and individual cancer cells to facilitate organized investigation of cancer-lymphocyte interactions in large patient cohorts [58]. A considerable number of studies have concluded that the location of immune cells in relation to malignant cells is of clinical value in different cancer types [60], [86], [87], [88], [89]. A higher concentration of CD8+ T-cells in remote stromal locations away from cancer cells independently predicted breast cancer-specific survival [88]. A higher concentration of CD3+ cells at the invasive margin of colorectal cancers significantly correlated with disease-free survival [60]. Also, higher CXCR5+ immune cell density at the invasive edge of different tumors correlated with bad outcome, contrary to a better good outcome if present central in the tumor [90]. Another study revealed that the presence of more dendritic cell aggregates in breast cancer correlated with a better prognosis [91].

Combination of genetic studies and digital pathology correlate with tumor clinical behavior

Malignant neoplasms as discussed before are extremely heterogeneous. Therefore, the expression of certain genes can differ significantly among different cell types in various regions within the same tumor. While OMICs provide high molecular resolution, they lack any spatial information. This lack of cellular/histological context can lead to bias/inconsistency of molecular data because of sampling issues. Conversely, histopathological features, in many instances, do not correlate accurately with the clinical behavior of tumors. Thus, integrating the histo-morphological context to genetic data may enable for a better understanding of the clinical behavior of malignant tumors.

For example, one study developed computational pipelines to integrate bioinformatics and digital pathology, aiming to subtype glioblastoma and renal cell carcinoma into biologically-relevant groups using TCGA data [61]. Another study used imaging and morphological parameters to identify three glioblastoma clusters characterized by different molecular events in histological areas enriched with specific microenvironmental cells [46]. A third study identified morphological patterns associated with survival outcome in triple-negative breast cancer [51]. In the latter study, nuclear features, regardless of their cell types, were quantitatively measured and correlated with gene expression profiling of the same set of triple-negative tumors. Some of these nuclear features, such as nuclear size, were found to be an independent prognostic value. Another breast cancer study found a strong correlation between analysis of nuclear morphology and copy number mutations [62]. Using manual annotation of 177 digitized frozen section images, Cooper et al. [25], defined the boundaries between necrosis and angiogenesis in 99 GBM tumors. Further analysis of the latter study revealed a correlation between the common defining histological features of GBM such as palisading necrosis and perivascular hyperplasia and the hypoxic upregulation of mesenchymal genes. Multiple other studies used TCGA resource to discover novel biological sub-classifications for a broad set of cancers [92], [93], [94], [95], [96], [97], [98].

Integrating 3D digital pathology and molecular genomics

The 3D imaging is not empirically novel in medicine as it has been implemented by non-invasive radiological imaging, which provided useful temporal and 3D images to neoplastic information for heterogenous primary and metastatic tumors [99], [100]. Histopathology, however, can offer a robust capability, not available to radiology, which is to map phenotypic and genetic aberrations at the cellular resolution. Traditionally, pathologic diagnosis using thin-cut sections mounted on glass slides made almost all tissue-based studies, including digitally scanned slides, limited to a two-dimensional (2D) space. Thus, image analysis of 2D microscopy was, until recently, subject to the same limitations of spatial information loss. The new advances in 3D microscopy allowed investigators to study disease-related biological structures in full authentic in situ 3D environments, which provided a wealth of information [101], [102].

In attempts to apply the analytical power of digital pathology in a 3D format, Clarke et al. [63] described a novel methodology of “3D pathology volumetric technique” to accurately measure tumor volumes in breast lumpectomies by preparing and processing whole mount slides and digitalizing them according to published methods [103], [104]. They used a pixel spacing of 2μm for better optimization of tumor detectability and computational feasibility [105]. Their 3D pathology volumetric method managed to improve the 3D visualizations of different tumor foci with better accuracy, than the conventional grossing techniques, of locating in situ breast carcinomas. A more recent study by the same group proposed an image-processing framework to incorporate the 3D digital pathology morphology with cell-specific molecular and biomarker features [106]. This incorporation of molecular biomarkers into the 3D image analytics offered additional advantages for the investigation of tumor clinical behavior at different spatial frames and a specific 3D micro-environmental context to the molecular alterations and signaling events.

Radio-histomics: potential clinical significance

Radio-histomics or the interface of radiology and histology is an emerging field. The integration of radiological imaging with digital pathological images, genomics and clinical data can provide a more holistic approach to understating and treating cancers. In a recent study, different parameters of MR imaging of prostatic tissues correlated significantly with certain histological features that differentiated malignant from normal tissue [64]. In addition to the merge of radio-histomics, digital pathology can benefit from the advanced image analysis technique used in radiology. A number of promising studies have already investigated the relation between the features of quantitative radiology imaging, genomic profiles and clinical outcomes [107], [108], [109], [110]. These quantitative radiology-imaging techniques can be of value if used within the realm of digital pathology. Further elaboration of the integration between digital pathology and other disciplines are illustrated in Figure 2.

Figure 2: The integration of clinical parameters with digital pathology image analysis, omics data and 3D radiological image analysis can provide a more comprehensive view that can be used for tumor subtyping, assessment of prognosis and predicting response to treatment.
Figure 2:

The integration of clinical parameters with digital pathology image analysis, omics data and 3D radiological image analysis can provide a more comprehensive view that can be used for tumor subtyping, assessment of prognosis and predicting response to treatment.

Challenges, potential solutions and future directions

The advances in digital slide scanning are not without technical challenges. Scanning a slide has to account for processing artifacts, staining variations between different slides, tissue section thickness and others. These present a considerable hurdle to embrace image analysis technology in clinical settings. However, these challenges are driving the evolution of digital pathology by inventing new methods to address them [35], [111], [112], [113], [114], [115]. Different algorithms have been developed to normalize pixel data and standardize color attributes in whole slide image database [37], [38], [112]. Other algorithmic approaches have been created to fix image defects due to tissue folding [35], [111], [113]. Currently, commercially available high-quality digital pathology solutions such as PerkinElmer (USA), TissueGnostics (Austria) and Definiens (Germany) are available. Thus far, a number of these methods were used to study tumor pathological characteristics in cancer specimens, leading to significant findings [39], [68], [84], [116], [117].

Other challenges that face digital pathology include its integration with other disciplines such as the molecular and bioinformatics disciplines. For example, one challenge for such integration is the hard choice between using frozen tissue or formalin-fixed and paraffin-embedded (FFPE) tissue. The quality of nucleic acids for molecular studies is generally better in frozen tissues. Thus, frozen tissues are often preferred for molecular profiling, while FFPE tissues are often used for pathological image analysis tools [39], [52], [53], [55], [56]. The future of addressing such a challenge remains open to discussion.

Another barrier to the progress of digital pathology is the lack of sharing and distribution of algorithms and picture processing applications beyond the reach of image analysis experts to the wider research society. Rendering free software and algorithms for defining specific features publicly accessible will promote the advancement in this field. Some shared platforms are currently available such as Pathology Analytic Imaging Standards (PAIS), which was created to support the standardization of image analysis algorithms and image features [118], [119]. Resources of web-based interfaces and online tools to support the visualization and analysis of pathology image data are becoming more readily available such as The Cancer Digital Slide Archive (CDSA, http://cancer.digitalslidearchive.net/) [120]. Another challenge for digital pathology is its need for independent validation of different findings, and the need to develop a standardized approach for assessment of correlations between morphology and molecular data. This is of particular importance to identify future genetic signatures, biomarkers and therapeutic targets that can be applied clinically.

Conclusions

The future advances in digital pathology are potentially limitless. Digital pathology represents not only a promising tool for disease diagnosis, but also for high-throughput and quantitative data extraction and analysis for multipurpose biomarker discovery for translational research on a wide spectrum of diseases [46], [121]. Thus, digital pathology can be considered critical to the future of precision medicine when treatment can be tailored to individual tumors in individual patients. While this new field holds great promise for better understanding of disease pathogenesis, progression and response to treatment, the scale of large imaging data can be overwhelming and technically challenging [41], [122]. However, with the advancement of computational hardware, software, online storages and the futuristic ideas of incorporating molecular and 3D information, it is almost impossible to predict the outcome except for its positivity.


Corresponding author: George M. Yousef, MD, PhD, FRCPC (Path), MSc, MBBCh, Department of Pediatric Laboratory Medicine, The Hospital for Sick Children, 555 University Avenue, Toronto, ON M5G 1X8, Canada; Department of Laboratory Medicine, and the Keenan Research Centre for Biomedical Science at the Li Ka Shing Knowledge Institute, St. Michael’s Hospital, Toronto, Canada; and Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada, Phone: +416-813-5990, Fax: +416-813-8674

  1. Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.

  2. Research funding: This work was supported by grants from the Kidney Cancer Research Network of Canada and the Canadian Urology Oncology group (KCRNC/CUOG) research trainee award, the Ontario Institute of Cancer Research (Transformative Pathology Fellowship award RFTP-004), and Ontario Molecular Pathology Research Network (OMPRN stream 1 grant #CPTRG-004).

  3. Employment or leadership: None declared.

  4. Honorarium: None declared.

  5. Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.

References

1. Nowell PC. The clonal evolution of tumor cell populations. Science 1976;194:23–8.10.1126/science.959840Search in Google Scholar PubMed

2. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;144:646–74.10.1016/j.cell.2011.02.013Search in Google Scholar PubMed

3. Casas-Selves M, Degregori J. How cancer shapes evolution, and how evolution shapes cancer. Evolution (N Y) 2011;4:624–34.10.1007/s12052-011-0373-ySearch in Google Scholar PubMed PubMed Central

4. Barsoum IB, Hamilton TK, Li X, Cotechini T, Miles EA, Siemens DR, et al. Hypoxia induces escape from innate immunity in cancer cells via increased expression of ADAM10: role of nitric oxide. Cancer Res 2011;71:7433–41.10.1158/0008-5472.CAN-11-2104Search in Google Scholar PubMed

5. Barsoum IB, Koti M, Siemens DR, Graham CH. Mechanisms of hypoxia-mediated immune escape in cancer. Cancer Res 2014;74:7185–90.10.1158/0008-5472.CAN-14-2598Search in Google Scholar PubMed

6. Barsoum IB, Smallwood CA, Siemens DR, Graham CH. A mechanism of hypoxia-mediated escape from adaptive immunity in cancer cells. Cancer Res 2014;74:665–74.10.1158/0008-5472.CAN-13-0992Search in Google Scholar PubMed

7. Black M, Barsoum IB, Truesdell P, Cotechini T, Macdonald-Goodfellow SK, Petroff M, et al. Activation of the PD-1/PD-L1 immune checkpoint confers tumor cell chemoresistance associated with increased metastasis. Oncotarget 2016;7:10557–67.10.18632/oncotarget.7235Search in Google Scholar PubMed PubMed Central

8. Graham C, Barsoum I, Kim J, Black M, Siemens RD. Mechanisms of hypoxia-induced immune escape in cancer and their regulation by nitric oxide. Redox Biol 2015;5:417.10.1016/j.redox.2015.09.022Search in Google Scholar PubMed

9. Calderwood SK. Tumor heterogeneity, clonal evolution, and therapy resistance: an opportunity for multitargeting therapy. Discov Med 2013;15:188–94.Search in Google Scholar

10. Lichner Z, Saleh C, Subramaniam V, Seivwright A, Prud’homme GJ, Yousef GM. miR-17 inhibition enhances the formation of kidney cancer spheres with stem cell/tumor initiating cell properties. Oncotarget 2015;6:5567–81.10.18632/oncotarget.1901Search in Google Scholar PubMed PubMed Central

11. Gillies RJ, Verduzco D, Gatenby RA. Evolutionary dynamics of carcinogenesis and why targeted therapy does not work. Nat Rev Cancer 2012;12:487–93.10.1038/nrc3298Search in Google Scholar PubMed PubMed Central

12. Chiche JD, Cariou A, Mira JP. Bench-to-bedside review: fulfilling promises of the Human Genome Project. Crit Care 2002;6:212–5.10.1186/cc1491Search in Google Scholar PubMed PubMed Central

13. Arsanious A, Bjarnason GA, Yousef GM. From bench to bedside: current and future applications of molecular profiling in renal cell carcinoma. Mol Cancer 2009;8:20.10.1186/1476-4598-8-20Search in Google Scholar PubMed PubMed Central

14. Gabril MY, Yousef GM. Informatics for practicing anatomical pathologists: marking a new era in pathology practice. Mod Pathol 2010;23:349–58.10.1038/modpathol.2009.190Search in Google Scholar PubMed

15. Pasic MD, Samaan S, Yousef GM. Genomic medicine: new frontiers and new challenges. Clin Chem 2013;59:158–67.10.1373/clinchem.2012.184622Search in Google Scholar PubMed

16. Ibrahim R, Pasic M, Yousef GM. Omics for personalized medicine: defining the current we swim in. Expert Rev Mol Diagn 2016;16:719–22.10.1586/14737159.2016.1164601Search in Google Scholar PubMed

17. Yousef GM. Personalized medicine in kidney cancer: learning how to walk before we run. Eur Urol 2015;68:1021–2.10.1016/j.eururo.2015.06.037Search in Google Scholar PubMed

18. Butz H, Szabo PM, Nofech-Mozes R, Rotondo F, Kovacs K, Mirham L, et al. Integrative bioinformatics analysis reveals new prognostic biomarkers of clear cell renal cell carcinoma. Clin Chem 2014;60:1314–26.10.1373/clinchem.2014.225854Search in Google Scholar PubMed

19. Girgis AH, Iakovlev VV, Beheshti B, Bayani J, Squire JA, Bui A, et al. Multilevel whole-genome analysis reveals candidate biomarkers in clear cell renal cell carcinoma. Cancer Res 2012;72:5273–84.10.1158/0008-5472.CAN-12-0656Search in Google Scholar PubMed

20. Kerr KM, Hirsch FR. Programmed death ligand-1 immunohistochemistry: friend or foe? Arch Pathol Lab Med 2016;140:326–31.10.5858/arpa.2015-0522-SASearch in Google Scholar PubMed

21. Lee KS, Lee K, Yun S, Moon S, Park Y, Han JH, et al. Prognostic relevance of programmed cell death ligand 1 expression in glioblastoma. J Neurooncol 2018;136:453–61.10.1007/s11060-017-2675-6Search in Google Scholar PubMed

22. Marginean EC, Melosky B. Is there a role for programmed death ligand-1 testing and immunotherapy in colorectal cancer with microsatellite instability? Part II – the challenge of programmed death ligand-1 testing and its role in microsatellite instability-high colorectal cancer. Arch Pathol Lab Med 2018;142:26–34.10.5858/arpa.2017-0041-RASearch in Google Scholar PubMed

23. Heindl A, Nawaz S, Yuan Y. Mapping spatial heterogeneity in the tumor microenvironment: a new era for digital pathology. Lab Invest 2015;95:377–84.10.1038/labinvest.2014.155Search in Google Scholar PubMed

24. van den Bent MJ. Interobserver variation of the histopathological diagnosis in clinical trials on glioma: a clinician’s perspective. Acta Neuropathol 2010;120:297–304.10.1007/s00401-010-0725-7Search in Google Scholar PubMed PubMed Central

25. Cooper LA, Kong J, Gutman DA, Dunn WD, Nalisnik M, Brat DJ. Novel genotype-phenotype associations in human cancers enabled by advanced molecular platforms and computational analysis of whole slide images. Lab Invest 2015;95:366–76.10.1038/labinvest.2014.153Search in Google Scholar PubMed PubMed Central

26. Kothari S, Phan JH, Stokes TH. Pathology imaging informatics for quantitative analysis of whole-slide images. J Am Med Inform Assoc 2013;20:1099–108.10.1136/amiajnl-2012-001540Search in Google Scholar PubMed PubMed Central

27. Cooper LA, Carter AB, Farris AB, Wang F, Kong J, Gutman DA, et al. Digital pathology: data-intensive frontier in medical imaging: health-information sharing, specifically of digital pathology, is the subject of this paper which discusses how sharing the rich images in pathology can stretch the capabilities of all otherwise well-practiced disciplines. Proc IEEE Inst Electr Electron Eng 2012;100:991–1003.10.1109/JPROC.2011.2182074Search in Google Scholar PubMed PubMed Central

28. Gurcan MN, Boucheron LE, Can A. Histopathological image analysis: a review. IEEE Rev Biomed Eng 2009;2:147–71.10.1109/RBME.2009.2034865Search in Google Scholar PubMed PubMed Central

29. Chang H, Han J, Borowsky A. Invariant delineation of nuclear architecture in glioblastoma multiforme for clinical and molecular association. IEEE Trans Med Imaging 2013;32:670–82.10.1109/TMI.2012.2231420Search in Google Scholar PubMed PubMed Central

30. Chang H, Nayak N, Spellman PT. Characterization of tissue histopathology via predictive sparse decomposition and spatial pyramid matching. Med Image Comput Comput Assist Interv 2013;16:91–8.10.1007/978-3-642-40763-5_12Search in Google Scholar PubMed PubMed Central

31. Hipp J, Smith SC, Cheng J, Tomlins SA. Optimization of complex cancer morphology detection using the SIVQ pattern recognition algorithm. Anal Cell Pathol (Amst) 2012;35:41–50.10.1155/2012/535819Search in Google Scholar

32. Hsu W, Markey MK, Wang MD. Biomedical imaging informatics in the era of precision medicine: progress, challenges, and opportunities. J Am Med Inform Assoc 2013;20:1010–3.10.1136/amiajnl-2013-002315Search in Google Scholar PubMed PubMed Central

33. Janowczyk A, Chandran S, Singh R. High-throughput biomarker segmentation on ovarian cancer tissue microarrays via hierarchical normalized cuts. IEEE Trans Biomed Eng 2012;59:1240–52.10.1109/TBME.2011.2179546Search in Google Scholar PubMed

34. Song Y, Treanor D, Bulpitt AJ. Unsupervised content classification based nonrigid registration of differently stained histology images. IEEE Trans Biomed Eng 2014;61:96–108.10.1109/TBME.2013.2277777Search in Google Scholar PubMed

35. Bautista PA, Yagi Y. Improving the visualization and detection of tissue folds in whole slide images through color enhancement. J Pathol Inform 2010;1:25.10.4103/2153-3539.73320Search in Google Scholar PubMed PubMed Central

36. Mosaliganti K, Janoos F, Irfanoglu O. Tensor classification of N-point correlation function features for histology tissue segmentation. Med Image Anal 2009;13:156–66.10.1016/j.media.2008.06.020Search in Google Scholar PubMed PubMed Central

37. Qi X, Xing F, Foran DJ. Robust segmentation of overlapping cells in histopathology specimens using parallel seed detection and repulsive level set. IEEE Trans Biomed Eng 2012;59:754–65.10.1109/TBME.2011.2179298Search in Google Scholar PubMed PubMed Central

38. Samsi S, Krishnamurthy AK, Gurcan MN. An efficient computational framework for the analysis of whole slide images: application to follicular lymphoma immunohistochemistry. J Comput Sci 2012;3:269–79.10.1016/j.jocs.2012.01.009Search in Google Scholar PubMed PubMed Central

39. Beck AH, Sangoi AR, Leung S. Systematic analysis of breast cancer morphology uncovers stromal features associated with survival. Sci Transl Med 2011;3:108ra113.10.1126/scitranslmed.3002564Search in Google Scholar PubMed

40. Mosaliganti K, Pan T, Ridgway R. An imaging workflow for characterizing phenotypical change in large histological mouse model datasets. J Biomed Inform 2008;41:863–73.10.1016/j.jbi.2008.03.013Search in Google Scholar PubMed PubMed Central

41. Kong J, Cooper LA, Wang F. Machine-based morphologic analysis of glioblastoma using whole-slide pathology images uncovers clinically relevant molecular correlates. PLoS One 2013;8:e81049.10.1371/journal.pone.0081049Search in Google Scholar PubMed PubMed Central

42. Kolin DL, Sy K, Rotondo F, Bassily MN, Kovacs K, Brezden-Masley C, et al. Prognostic significance of human tissue kallikrein-related peptidases 6 and 10 in gastric cancer. Biol Chem 2014;395:1087–93.10.1515/hsz-2014-0143Search in Google Scholar PubMed

43. Scopa CD, Melachrinou M, Saradopoulou C, Merino MJ. The significance of the grooved nucleus in thyroid lesions. Mod Pathol 1993;6:691–4.Search in Google Scholar

44. Samaratunga H, Gianduzzo T, Delahunt B. The ISUP system of staging, grading and classification of renal cell neoplasia. J Kidney Cancer VHL 2014;1:26–39.10.15586/jkcvhl.2014.11Search in Google Scholar PubMed PubMed Central

45. Appin CL, Gao J, Chisolm C. Glioblastoma with oligodendroglioma component (GBM-O): molecular genetic and clinical characteristics. Brain Pathol 2013;23:454–61.10.1111/bpa.12018Search in Google Scholar PubMed PubMed Central

46. Cooper LA, Kong J, Gutman DA. Integrated morphologic analysis for the identification and characterization of disease subtypes. J Am Med Inform Assoc 2012;19:317–23.10.1136/amiajnl-2011-000700Search in Google Scholar PubMed PubMed Central

47. Gupta M, Djalilvand A, Brat DJ. Clarifying the diffuse gliomas: an update on the morphologic features and markers that discriminate oligodendroglioma from astrocytoma. Am J Clin Pathol 2005;124:755–68.10.1309/6JNX4PA60TQ5U5VGSearch in Google Scholar

48. Hegi ME, Janzer RC, Lambiv WL. Presence of an oligodendroglioma-like component in newly diagnosed glioblastoma identifies a pathogenetically heterogeneous subgroup and lacks prognostic value: central pathology review of the EORTC_26981/NCIC_CE.3 trial. Acta Neuropathol 2012;123:841–52.10.1007/s00401-011-0938-4Search in Google Scholar PubMed

49. Peraud A, Watanabe K, Schwechheimer K. Genetic profile of the giant cell glioblastoma. Lab Invest 1999;79:123–9.Search in Google Scholar

50. Perry A, Aldape KD, George DH. Small cell astrocytoma: an aggressive variant that is clinicopathologically and genetically distinct from anaplastic oligodendroglioma. Cancer 2004;101:2318–26.10.1002/cncr.20625Search in Google Scholar PubMed

51. Wang C, Pecot T, Zynger DL. Identifying survival associated morphological features of triple negative breast cancer using multiple datasets. J Am Med Inform Assoc 2013;20:680–7.10.1136/amiajnl-2012-001538Search in Google Scholar PubMed PubMed Central

52. Doyle S, Feldman M, Tomaszewski J. A boosted Bayesian multiresolution classifier for prostate cancer detection from digitized needle biopsies. IEEE Trans Biomed Eng 2012;59:1205–18.10.1109/TBME.2010.2053540Search in Google Scholar PubMed

53. Tabesh A, Teverovskiy M, Pang HY. Multifeature prostate cancer diagnosis and Gleason grading of histological images. IEEE Trans Med Imaging 2007;26:1366–78.10.1109/TMI.2007.898536Search in Google Scholar PubMed

54. Khan AM, Eldaly H, Rajpoot NM. A gamma-Gaussian mixture model for detection of mitotic cells in breast cancer histopathology images. J Pathol Inform 2013;4:11.10.4103/2153-3539.112696Search in Google Scholar PubMed PubMed Central

55. Roux L, Racoceanu D, Lomenie N. Mitosis detection in breast cancer histological images An ICPR 2012 contest. J Pathol Inform 2013;4:8.10.4103/2153-3539.112693Search in Google Scholar PubMed PubMed Central

56. Lewis JS Jr., Ali S, Luo J. A quantitative histomorphometric classifier (QuHbIC) identifies aggressive versus indolent p16-positive oropharyngeal squamous cell carcinoma. Am J Surg Pathol 2014;38:128–37.10.1097/PAS.0000000000000086Search in Google Scholar PubMed PubMed Central

57. Kolin DL, Sy K, Rotondo F, Bassily MN, Kovacs K, Brezden-Masley C, et al. Prognostic significance of human tissue kallikrein-related peptidases 11 and 15 in gastric cancer. Tumour Biol 2016;37:437–46.10.1007/s13277-015-3802-7Search in Google Scholar PubMed

58. Yuan Y. Modelling the spatial and molecular heterogeneity of lymphocytic infiltration in triple-negative breast cancer. J R Soc Interface 2014;12:20141153.10.1098/rsif.2014.1153Search in Google Scholar PubMed PubMed Central

59. Feichtenbeiner A, Haas M, Buttner M. Critical role of spatial interaction between CD8(+) and Foxp3(+) cells in human gastric cancer: the distance matters. Cancer Immunol Immunother 2014;63:111–9.10.1007/s00262-013-1491-xSearch in Google Scholar PubMed

60. Galon J, Costes A, Sanchez-Cabo F. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science 2006;313:1960–4.10.1126/science.1129139Search in Google Scholar PubMed

61. Chang H, Fontenay GV, Han J. Morphometic analysis of TCGA glioblastoma multiforme. BMC Bioinformatics 2011;12:484.10.1186/1471-2105-12-484Search in Google Scholar PubMed PubMed Central

62. Yuan Y, Failmezger H, Rueda OM. Quantitative image analysis of cellular heterogeneity in breast tumors complements genomic profiling. Sci Transl Med 2012;4.10.1126/scitranslmed.3004330Search in Google Scholar PubMed

63. Clarke GM, Murray M, Holloway CM, Liu K, Zubovits JT, Yaffe MJ. 3D pathology volumetric technique: a method for calculating breast tumour volume from whole-mount serial section images. Int J Breast Cancer 2012;2012:9.10.1155/2012/691205Search in Google Scholar PubMed PubMed Central

64. Langer DL, van der Kwast TH, Evans AJ, Plotkin A, Trachtenberg J, Wilson BC, et al. Prostate tissue composition and MR measurements: investigating the relationships between ADC, T2, K(trans), v(e), and corresponding histologic features. Radiology 2010;255:485–94.10.1148/radiol.10091343Search in Google Scholar PubMed

65. Hanahan D, Coussens LM. Accessories to the crime: functions of cells recruited to the tumor microenvironment. Cancer Cell 2012;21:309–22.10.1016/j.ccr.2012.02.022Search in Google Scholar PubMed

66. Junttila MR, de Sauvage FJ. Influence of tumour micro-environment heterogeneity on therapeutic response. Nature 2013;501:346–54.10.1038/nature12626Search in Google Scholar PubMed

67. Paolino M, Choidas A, Wallner S. The E3 ligase Cbl-b and TAM receptors regulate cancer metastasis via natural killer cells. Nature 2014;507:508–12.10.1038/nature12998Search in Google Scholar PubMed PubMed Central

68. Tan KL, Scott DW, Hong F. Tumor-associated macrophages predict inferior outcomes in classic Hodgkin lymphoma: a correlative study from the E2496 Intergroup trial. Blood 2012;120:3280–7.10.1182/blood-2012-04-421057Search in Google Scholar PubMed PubMed Central

69. Elkabets M, Gifford AM, Scheel C. Human tumors instigate granulin-expressing hematopoietic cells that promote malignancy by activating stromal fibroblasts in mice. J Clin Invest 2011;121:784–99.10.1172/JCI43757Search in Google Scholar PubMed PubMed Central

70. Wille C, Kohler C, Armacki M. Protein kinase D2 induces invasion of pancreatic cancer cells by regulating matrix metalloproteinases. Mol Biol Cell 2014;25:324–36.10.1091/mbc.e13-06-0334Search in Google Scholar

71. Goddard JC, Sutton CD, Furness PN. A computer image analysis system for microvessel density measurement in solid tumours. Angiogenesis 2002;5:15–20.10.1023/A:1021518315757Search in Google Scholar

72. Kim NT, Elie N, Plancoulaine B. An original approach for quantification of blood vessels on the whole tumour section. Anal Cell Pathol 2003;25:63–75.10.1155/2003/473902Search in Google Scholar PubMed PubMed Central

73. Mikalsen LT, Dhakal HP, Bruland OS. Quantification of angiogenesis in breast cancer by automated vessel identification in CD34 immunohistochemical sections. Anticancer Res 2011;31:4053–60.Search in Google Scholar

74. Fridman WH, Pages F, Sautes-Fridman C. The immune contexture in human tumours: impact on clinical outcome. Nat Rev Cancer 2012;12:298–306.10.1038/nrc3245Search in Google Scholar PubMed

75. Galon J, Mlecnik B, Bindea G. Towards the introduction of the ‘Immunoscore’ in the classification of malignant tumours. J Pathol 2014;232:199–209.10.1002/path.4287Search in Google Scholar PubMed PubMed Central

76. Ascierto ML, Kmieciak M, Idowu MO. A signature of immune function genes associated with recurrence-free survival in breast cancer patients. Breast Cancer Res Treat 2012;131:871–80.10.1007/s10549-011-1470-xSearch in Google Scholar PubMed PubMed Central

77. Calabro A, Beissbarth T, Kuner R. Effects of infiltrating lymphocytes and estrogen receptor on gene expression and prognosis in breast cancer. Breast Cancer Res Treat 2009;116:69–77.10.1007/s10549-008-0105-3Search in Google Scholar PubMed

78. Gu-Trantien C, Loi S, Garaud S. CD4(+) follicular helper T cell infiltration predicts breast cancer survival. J Clin Invest 2013;123:2873–92.10.1172/JCI67428Search in Google Scholar PubMed PubMed Central

79. Rody A, Holtrich U, Pusztai L. T-cell metagene predicts a favorable prognosis in estrogen receptor-negative and HER2-positive breast cancers. Breast Cancer Res 2009;11:R15.10.1186/bcr2234Search in Google Scholar PubMed PubMed Central

80. Andre F, Dieci MV, Dubsky P. Molecular pathways: involvement of immune pathways in the therapeutic response and outcome in breast cancer. Clin Cancer Res 2013;19:28–33.10.1158/1078-0432.CCR-11-2701Search in Google Scholar PubMed

81. Basavanhally AN, Ganesan S, Agner S. Computerized image-based detection and grading of lymphocytic infiltration in HER2+ breast cancer histopathology. IEEE Trans Biomed Eng 2010;57:642–53.10.1109/TBME.2009.2035305Search in Google Scholar PubMed

82. Janowczyk A, Chandran S, Madabhushi A. Quantifying local heterogeneity via morphologic scale: Distinguishing tumoral from stromal regions. J Pathol Inform 2013;4:S8.10.4103/2153-3539.109865Search in Google Scholar PubMed PubMed Central

83. Setiadi AF, Ray NC, Kohrt HE. Quantitative, architectural analysis of immune cell subsets in tumor-draining lymph nodes from breast cancer patients and healthy lymph nodes. PLoS One 2010;5:e12420.10.1371/journal.pone.0012420Search in Google Scholar PubMed PubMed Central

84. Kruger JM, Wemmert C, Sternberger L. Combat or surveillance? Evaluation of the heterogeneous inflammatory breast cancer microenvironment. J Pathol 2013;229:569–78.10.1002/path.4150Search in Google Scholar PubMed

85. Mansfield J, Nelson L, van der Loos C. Abstract C286: phenotyping TILs in situ: automated enumeration of Tregs and Tacts in solid tumors. Mol Cancer Ther 2013;12:C286.10.1158/1535-7163.TARG-13-C286Search in Google Scholar

86. Denkert C, Loibl S, Noske A. Tumor-associated lymphocytes as an independent predictor of response to neoadjuvant chemotherapy in breast cancer. J Clin Oncol 2010;28:105–13.10.1200/JCO.2009.23.7370Search in Google Scholar PubMed

87. Loi S, Sirtaine N, Piette F. Prognostic and predictive value of tumor-infiltrating lymphocytes in a phase III randomized adjuvant breast cancer trial in node-positive breast cancer comparing the addition of docetaxel to doxorubicin with doxorubicin-based chemotherapy: BIG 02-98. J Clin Oncol 2013;31:860–7.10.1200/JCO.2011.41.0902Search in Google Scholar PubMed

88. Mahmoud SM, Paish EC, Powe DG. Tumor-infiltrating CD8+ lymphocytes predict clinical outcome in breast cancer. J Clin Oncol 2011;29:1949–55.10.1200/JCO.2010.30.5037Search in Google Scholar PubMed

89. Zhang L, Conejo-Garcia JR, Katsaros D. Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N Engl J Med 2003;348:203–13.10.1056/NEJMoa020177Search in Google Scholar PubMed

90. Bindea G, Mlecnik B, Tosolini M. Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer. Immunity 2013;39:782–95.10.1016/j.immuni.2013.10.003Search in Google Scholar PubMed

91. Chang AY, Bhattacharya N, Mu J. Spatial organization of dendritic cells within tumor draining lymph nodes impacts clinical outcome in breast cancer patients. J Transl Med 2013;11:242.10.1186/1479-5876-11-242Search in Google Scholar PubMed PubMed Central

92. Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian carcinoma. Nature 2011;474:609–15.10.1038/nature10166Search in Google Scholar PubMed PubMed Central

93. Cancer Genome Atlas Network. Comprehensive molecular portraits of human breast tumours. Nature 2012;490:61–70.10.1038/nature11412Search in Google Scholar PubMed PubMed Central

94. Cancer Genome Atlas Network. Comprehensive molecular characterization of human colon and rectal cancer. Nature 2012;487:330–7.10.1038/nature11252Search in Google Scholar PubMed PubMed Central

95. Cancer Genome Atlas Research Network. Comprehensive genomic characterization of squamous cell lung cancers. Nature 2012;489:519–25.10.1038/nature11404Search in Google Scholar PubMed PubMed Central

96. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of clear cell renal cell carcinoma. Nature 2013;499:43–9.10.1038/nature12222Search in Google Scholar PubMed PubMed Central

97. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of urothelial bladder carcinoma. Nature 2014;507:315–22.10.1038/nature12965Search in Google Scholar PubMed PubMed Central

98. Rutledge WC, Kong J, Gao J. Tumor-infiltrating lymphocytes in glioblastoma are associated with specific genomic alterations and related to transcriptional class. Clin Cancer Res 2013;19:4951–60.10.1158/1078-0432.CCR-13-0551Search in Google Scholar PubMed PubMed Central

99. Davnall F, Yip CS, Ljungqvist G. Assessment of tumor heterogeneity: an emerging imaging tool for clinical practice? Insights Imaging 2012;3:573–89.10.1007/s13244-012-0196-6Search in Google Scholar PubMed PubMed Central

100. Ganeshan B, Miles KA. Quantifying tumour heterogeneity with CT. Cancer Imaging 2013;13:140–9.10.1102/1470-7330.2013.0015Search in Google Scholar PubMed PubMed Central

101. Liang Y, Wang F, Treanor D, Magee D, Roberts N, Teodoro G, et al. A framework for 3D vessel analysis using whole slide images of liver tissue sections. Int J Comput Biol Drug Des 2016;9:102–19.10.1504/IJCBDD.2016.074983Search in Google Scholar PubMed PubMed Central

102. Roberts N, Magee D, Song Y, Brabazon K, Shires M, Crellin D, et al. Toward routine use of 3D histopathology as a research tool. Am J Pathol 2012;180:1835–42.10.1016/j.ajpath.2012.01.033Search in Google Scholar PubMed PubMed Central

103. Tot T, Pekár G, Hofmeyer S, Sollie T, Gere M, Tarján M. The distribution of lesions in 1–14-mm invasive breast carcinomas and its relation to metastatic potential. Virchows Archiv 2009;455:109–15.10.1007/s00428-009-0808-9Search in Google Scholar PubMed

104. Clarke GM, Peressotti C, Constantinou P, Hosseinzadeh D, Martel A, Yaffe MJ. Increasing specimen coverage using digital whole-mount breast pathology: implementation, clinical feasibility and application in research. Comput Med Imaging Graph 2011;35:531–41.10.1016/j.compmedimag.2011.05.002Search in Google Scholar PubMed

105. Clarke GM, Zubovits JT, Katic M, Peressotti C, Yaffe MJ. Spatial resolution requirements for acquisition of the virtual screening slide for digital whole-specimen breast histopathology. Human Pathol 2007;38:1764–71.10.1016/j.humpath.2007.04.006Search in Google Scholar PubMed

106. Liang Y, Wang F, Zhang P, Saltz JH, Brat DJ, Kong J. Development of a framework for large scale three-dimensional pathology and biomarker imaging and spatial analytics. AMIA Jt Summits Transl Sci Proc 2017;2017:75–84.Search in Google Scholar

107. Gevaert O, Xu J, Hoang CD. Non-small cell lung cancer: identifying prognostic imaging biomarkers by leveraging public gene expression microarray data[mdash]methods and preliminary results. Radiology 2012;264:387–96.10.1148/radiol.12111607Search in Google Scholar PubMed PubMed Central

108. Gutman DA, Cooper LA, Hwang SN. MR imaging predictors of molecular profile and survival: multi-institutional study of the TCGA glioblastoma data set. Radiology 2013;267:560–9.10.1148/radiol.13120118Search in Google Scholar PubMed PubMed Central

109. Jain R, Poisson L, Narang J. Genomic mapping and survival prediction in glioblastoma: molecular subclassification strengthened by hemodynamic imaging biomarkers. Radiology 2013;267:212–20.10.1148/radiol.12120846Search in Google Scholar PubMed PubMed Central

110. Zinn PO, Mahajan B, Sathyan P. Radiogenomic mapping of edema/cellular invasion MRI-phenotypes in glioblastoma multiforme. PLoS One 2011;6:e25451.10.1371/journal.pone.0025451Search in Google Scholar PubMed PubMed Central

111. Bautista PA, Hashimoto N, Yagi Y. Color standardization in whole slide imaging using a color calibration slide. J Pathol Inform 2014;5:4.10.4103/2153-3539.126153Search in Google Scholar PubMed PubMed Central

112. Khan AM, Rajpoot N, Treanor D. A nonlinear mapping approach to stain normalization in digital histopathology images using image-specific color deconvolution. IEEE Trans Biomed Eng 2014;61:1729–38.10.1109/TBME.2014.2303294Search in Google Scholar PubMed

113. Murakami Y, Abe T, Hashiguchi A. Color correction for automatic fibrosis quantification in liver biopsy specimens. J Pathol Inform 2013;4:36.10.4103/2153-3539.124009Search in Google Scholar PubMed PubMed Central

114. Chappelow J, Tomaszewski JE, Feldman M. HistoStitcher((c)): an interactive program for accurate and rapid reconstruction of digitized whole histological sections from tissue fragments. Comput Med Imaging Graph 2011;35:557–67.10.1016/j.compmedimag.2011.01.010Search in Google Scholar PubMed PubMed Central

115. Kothari S, Phan JH, Wang MD. Eliminating tissue-fold artifacts in histopathological whole-slide images for improved image-based prediction of cancer grade. J Pathol Inform 2013;4:22.10.4103/2153-3539.117448Search in Google Scholar PubMed PubMed Central

116. Tuominen VJ, Ruotoistenmaki S, Viitanen A. ImmunoRatio: a publicly available web application for quantitative image analysis of estrogen receptor (ER), progesterone receptor (PR), and Ki-67. Breast Cancer Res 2010;12:R56.10.1186/bcr2615Search in Google Scholar PubMed PubMed Central

117. Varghese F, Bukhari AB, Malhotra R. IHC Profiler: an open source plugin for the quantitative evaluation and automated scoring of immunohistochemistry images of human tissue samples. PLoS One 2014;9:e96801.10.1371/journal.pone.0096801Search in Google Scholar PubMed PubMed Central

118. Wang F, Kong J, Cooper L. A data model and database for high-resolution pathology analytical image informatics. J Pathol Inform 2011;2:32.10.4103/2153-3539.83192Search in Google Scholar PubMed PubMed Central

119. Wang F, Kong J, Gao J. A high-performance spatial database based approach for pathology imaging algorithm evaluation. J Pathol Inform 2013;4:5.10.4103/2153-3539.108543Search in Google Scholar PubMed PubMed Central

120. Gutman DA, Cobb J, Somanna D. Cancer digital slide archive: an informatics resource to support integrated in silico analysis of TCGA pathology data. J Am Med Inform Assoc 2013;20:1091–8.10.1136/amiajnl-2012-001469Search in Google Scholar PubMed PubMed Central

121. Kong J, Cooper LA, Wang F, Gutman DA, Gao J, Chisolm C, et al. Integrative, multimodal analysis of glioblastoma using TCGA molecular data, pathology images, and clinical outcomes. IEEE Trans Biomed Eng 2011;58:3469–74.10.1109/TBME.2011.2169256Search in Google Scholar PubMed PubMed Central

122. Jara-Lazaro AR, Thamboo TP, Teh M, Tan PH. Digital pathology: exploring its applications in diagnostic surgical pathology practice. Pathology 2010;42:512–8.10.3109/00313025.2010.508787Search in Google Scholar PubMed

Received: 2018-07-25
Accepted: 2018-09-28
Published Online: 2018-12-20
Published in Print: 2019-08-27

©2019 Walter de Gruyter GmbH, Berlin/Boston

Downloaded on 28.3.2024 from https://www.degruyter.com/document/doi/10.1515/dx-2018-0064/html
Scroll to top button