Search

Abstract Understanding factors that impact prognosis for cancer patients have high clinical relevance for treatment decisions and monitoring of the disease outcome. Advances in artificial intelligence (AI) and digital pathology offer an exciting opportunity to capitalize on the use of whole slide images (WSIs) of hematoxylin and eosin (H&E) stained tumor tissue for objective prognosis and prediction of response to targeted therapies. AI models often require hand-delineated annotations for effective training which may not be readily available for larger data sets. In this study, we investigated whether AI models can be trained without region-level annotations and solely on patient-level survival data. We present a weakly supervised survival convolutional neural network (WSS-CNN) approach equipped with a visual attention mechanism for predicting overall survival. The inclusion of visual attention provides insights into regions of the tumor microenvironment with the pathological interpretation which may improve our understanding of the disease pathomechanism. We performed this analysis on two independent, multi-center patient data sets of lung (which is publicly available data) and bladder urothelial carcinoma. We perform univariable and multivariable analysis and show that WSS-CNN features are prognostic of overall survival in both tumor indications. The presented results highlight the significance of computational pathology algorithms for predicting prognosis using H&E stained images alone and underpin the use of computational methods to improve the efficiency of clinical trial studies.

For personalization of medicine, increasingly clinical and demographic data are integrated into nomograms for prognostic use, while molecular biomarkers are being developed to add independent diagnostic, prognostic, or management information. In a number of cases in surgical pathology, morphometric quantitation is already performed manually or semi-quantitatively, with this effort contributing to diagnostic workup. Digital whole slide imaging, coupled with emerging image analysis algorithms, offers great promise as an adjunctive tool for the surgical pathologist in areas of screening, quality assurance, consistency, and quantitation. We have recently reported such an algorithm, SIVQ (Spatially Invariant Vector Quantization), which avails itself of the geometric advantages of ring vectors for pattern matching, and have proposed a number of potential applications. One key test, however, remains the need for demonstration and optimization of SIVQ for discrimination between foreground (neoplasm- malignant epithelium) and background (normal parenchyma, stroma, vessels, inflammatory cells). Especially important is the determination of relative contributions of each key SIVQ matching parameter with respect to the algorithm’s overall detection performance. Herein, by combinatorial testing of SIVQ ring size, sub-ring number, and inter-ring wobble parameters, in the setting of a morphologically complex bladder cancer use case, we ascertain the relative contributions of each of these parameters towards overall detection optimization using urothelial carcinoma as a use case, providing an exemplar by which this algorithm and future histology-oriented pattern matching tools may be validated and subsequently, implemented broadly in other appropriate microscopic classification settings.

Introduction: The advent of digital slides offers new opportunities within the practice of pathology such as the use of image analysis techniques to facilitate computer aided diagnosis (CAD) solutions. Use of CAD holds promise to enable new levels of decision support and allow for additional layers of quality assurance and consistency in rendered diagnoses. However, the development and testing of prostate cancer CAD solutions requires a ground truth map of the cancer to enable the generation of receiver operator characteristic (ROC) curves. This requires a pathologist to annotate, or paint, each of the malignant glands in prostate cancer with an image editor software - a time consuming and exhaustive process.

Recently, with the advent of the 7th edition of the AJCC Cancer Staging manual, at least one set of criteria (e.g. breast) were modified to now require the measurement of maximal depth of stromal invasion. With the current manual interpretive morphological approaches typically employed by surgical pathologists to assess tumor extent, the specialty now potentially has stumbled upon a crossroads of practice, where the diagnostic criteria have exceeded the capabilities of our commonly available tools. While whole slide imaging (WSI) technology holds the potential to offer many improvements in clinical workflow over conventional slide microscopy including unambiguous utility for facilitating quantitative diagnostic tasks with one important example being the determination of both linear dimension and surface area. However, the availability of histology data in digital form is of little utility if time-consuming and cumbersome manual workflow steps are necessarily imposed upon the pathologist in order to generate such measurements, especially as encountered with the complex and ill-defined shapes inherent to infiltrative tumors. In this communication, we demonstrate the utility of the recently described SIVQ algorithm to serve as the basis of a highly accurate, precise and semi-automated tool for direct surface area measurement of tumor infiltration from WSI data sets. By anticipating the current trend in cancer staging that emphasizes increasingly precise feature characterization, as witnessed by the recent publication of AJCC's 7th edition of the Cancer Staging Manual, this tool holds promise to will be of value to pathologists for clinical utility.

Introduction: Laser capture microdissection (LCM) facilitates procurement of defined cell populations for study in the context of histopathology. The morphologic assessment step in the LCM procedure is time consuming and tedious, thus restricting the utility of the technology for large applications. Results: Here, we describe the use of Spatially Invariant Vector Quantization (SIVQ) for histological analysis and LCM. Using SIVQ, we selected vectors as morphologic predicates that were representative of normal epithelial or cancer cells and then searched for phenotypically similar cells across entire tissue sections. The selected cells were subsequently auto-microdissected and the recovered RNA was analyzed by expression microarray. Gene expression profiles from SIVQ-LCM and standard LCM-derived samples demonstrated highly congruous signatures, confirming the equivalence of the differing microdissection methods. Conclusion: SIVQ-LCM improves the work-flow of microdissection in two significant ways. First, the process is transformative in that it shifts the pathologist′s role from technical execution of the entire microdissection to a limited-contact supervisory role, enabling large-scale extraction of tissue by expediting subsequent semi-autonomous identification of target cell populations. Second, this work-flow model provides an opportunity to systematically identify highly constrained cell populations and morphologically consistent regions within tissue sections. Integrating SIVQ with LCM in a single environment provides advanced capabilities for efficient and high-throughput histological-based molecular studies.

Introduction: Spatially invariant vector quantization (SIVQ) is a texture and color-based image matching algorithm that queries the image space through the use of ring vectors. In prior studies, the selection of one or more optimal vectors for a particular feature of interest required a manual process, with the user initially stochastically selecting candidate vectors and subsequently testing them upon other regions of the image to verify the vector′s sensitivity and specificity properties (typically by reviewing a resultant heat map). In carrying out the prior efforts, the SIVQ algorithm was noted to exhibit highly scalable computational properties, where each region of analysis can take place independently of others, making a compelling case for the exploration of its deployment on high-throughput computing platforms, with the hypothesis that such an exercise will result in performance gains that scale linearly with increasing processor count. Methods: An automated process was developed for the selection of optimal ring vectors to serve as the predicate matching operator in defining histopathological features of interest. Briefly, candidate vectors were generated from every possible coordinate origin within a user-defined vector selection area (VSA) and subsequently compared against user-identified positive and negative "ground truth" regions on the same image. Each vector from the VSA was assessed for its goodness-of-fit to both the positive and negative areas via the use of the receiver operating characteristic (ROC) transfer function, with each assessment resulting in an associated area-under-the-curve (AUC) figure of merit. Results: Use of the above-mentioned automated vector selection process was demonstrated in two cases of use: First, to identify malignant colonic epithelium, and second, to identify soft tissue sarcoma. For both examples, a very satisfactory optimized vector was identified, as defined by the AUC metric. Finally, as an additional effort directed towards attaining high-throughput capability for the SIVQ algorithm, we demonstrated the successful incorporation of it with the MATrix LABoratory (MATLAB TM ) application interface. Conclusion: The SIVQ algorithm is suitable for automated vector selection settings and high throughput computation.

Introduction: The increasing availability of whole slide imaging (WSI) data sets (digital slides) from glass slides offers new opportunities for the development of computer-aided diagnostic (CAD) algorithms. With the all-digital pathology workflow that these data sets will enable in the near future, literally millions of digital slides will be generated and stored. Consequently, the field in general and pathologists, specifically, will need tools to help extract actionable information from this new and vast collective repository. Methods: To address this limitation, we designed and implemented a tool (dCORE) to enable the systematic capture of image tiles with constrained size and resolution that contain desired histopathologic features. Results: In this communication, we describe a user-friendly tool that will enable pathologists to mine digital slides archives to create image microarrays (IMAs). IMAs are to digital slides as tissue microarrays (TMAs) are to cell blocks. Thus, a single digital slide could be transformed into an array of hundreds to thousands of high quality digital images, with each containing key diagnostic morphologies and appropriate controls. Current manual digital image cut-and-paste methods that allow for the creation of a grid of images (such as an IMA) of matching resolutions are tedious. Conclusion: The ability to create IMAs representing hundreds to thousands of vetted morphologic features has numerous applications in education, proficiency testing, consensus case review, and research. Lastly, in a manner analogous to the way conventional TMA technology has significantly accelerated in situ studies of tissue specimens use of IMAs has similar potential to significantly accelerate CAD algorithm development.