The COVID-19 Community Research Partnership: a multistate surveillance platform for characterizing the epidemiology of the SARS-CoV-2 pandemic

The COVID-19 Community Research Partnership (CCRP) is a multisite surveillance platform designed to characterize the epidemiology of the Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-COV-2) pandemic. This article describes the CCRP study design and methodology. The CCRP includes two prospective cohorts, one with six health systems in the mid-Atlantic and southern USA, and the other with six health systems in North Carolina. With enrollment beginning in April 2020, sites invited persons within their healthcare systems as well as community members to participate in daily surveillance for symptoms of COVID-like illnesses, testing, and risk behaviors. Participants with electronic health records (EHRs) were also asked to volunteer data access. Subsets of participants, representative of the general population and including oversampling of populations of interest, were selected for repeated at-home serology testing. By October 2021, 65 739 participants (62 261 adult and 3478 pediatric) were enrolled, with 89% providing syndromic data, 74% providing EHR data, and 70% participating in one of the two serology sub-studies. An average of 62% of the participants completed a daily survey at least once a week, and 55% of the serology kits were returned. The CCRP provides rich regional epidemiologic data and the opportunity to more fully characterize the risks and sequelae of SARS-CoV-2 infection.

AB1305 A SYSTEMATIC REVIEW OF AA AMYLOIDOSIS AMONG PATIENTS WITH BEHÇET’S SYNDROME

Background

Data on patients with Behçet’s syndrome (BS) complicated with AA amyloidosis is limited to case reports or case series with a small number of patients.

Objectives

In this study, we aimed to perform a systematic review (SR) of published reports on BS patients with AA amyloidosis.

Methods

PubMed and EMBASE were searched with the keywords “Behcet* AND amyloidosis”, without date and language restriction, until May 2020. Two independent reviewers (SNE, GK) performed title/abstract and full text screening and data extraction. A third reviewer (GH) made the final decision in case of disagreement between the two reviewers. Studies that reported patients who were reported by authors as having BS and AA amyloidosis were included. The risk of bias assessment was done using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) tool.

Results

The systematic literature search yielded 760 articles of which 703 were excluded after title and abstract review. After full-text review, we further excluded 15 duplicate articles and 1 article was added after handsearching the reference lists of the full texts. Finally, we included 43 articles reporting 96 cases. Among these articles, 38 were case reports and 5 were case series reporting between 6 and 14 patients. All patients but 8 were reported from Mediterranean countries. The quality of all articles according to GRADE was very low due to the lack of a control group.

The main features of the patients were male predominance (81/96, 84%), a high frequency of major organ involvement (62/80, 77.5%) especially vascular involvement (60%), a low frequency of comorbidities predisposing to AA amyloidosis (11/96, 11.5%), and a very low frequency of gastrointestinal involvement (3/72, 4%). All but 8 patients were diagnosed with BS and AA amyloidosis simultaneously. The most common presentation was nephrotic syndrome (60/81, 74%). Presenting symptoms other than proteinuria were diarrhea (n=2), acute renal failure (n=2), upper gastrointestinal bleeding (n=1), end stage renal disease (ESRD) (n=1), cardiac symptoms due to cor pulmonale (n=1), and hypertension (n=1). Renal biopsy (72%) and rectal biopsy (17%) were the most commonly used procedures to diagnose AA amyloidosis.

After diagnosing AA amyloidosis, colchicine was initiated in 58 patients, cyclophosphamide in 16, and biologics in 3 (1 anakinra and 2 tocilizumab). In the 67 patients with available data on follow-up, 43% of the patients were followed-up for ≤1 year and median follow-up duration was 20 months (IQR: 4-48). Among the 64 patients with available data, 30 (47%) had developed ESRD. Among the 72 patients with available data on survival status, 30 patients (42%) had died. Ten patients (33%) had died within 6 months, 15 had died after a median follow-up of 48 months (IQR: 24-150), and follow-up duration was not available in the remaining 5 patients including 3 patients whose diagnoses were made by autopsy. Reasons for death were infection (n=7), ESRD (n=6), intractable diarrhea (n=3), pulmonary embolism (n=1), cor pulmonale (n=1), hemorrhage due to pulmonary artery aneurysm (n=1), liver cirrhosis (n=1), gastric cancer (n=1), subarachnoid hemorrhage (n=1), and not reported (n=8).

Conclusion

Male gender and major organ involvement, especially vascular involvement, appear to be risk factors for the development of AA amyloidosis in BS patients. While BS patients complicated with AA amyloidosis have been reported rarely, it is a fatal complication of BS. One third of the patients had died within 6 months after AA amyloidosis diagnosis.

Disclosure of Interests

Güzin Karatemiz: None declared, Sinem Nihal Esatoglu Speakers bureau: Sinem Nihal Esatoglu has received honorariums for presentations from UCB Pharma, Roche, Pfizer, and Merck Sharp Dohme., Mert Gurcan: None declared, Yesim Ozguler Speakers bureau: Yesim Ozguler has received honorariums for presentations from UCB Pharma, Novartis, and Pfizer., Sebahattin Yurdakul: None declared, Vedat Hamuryudan Speakers bureau: Vedat Hamuryudan has served as a speaker for AbbVie, Celgene, Novartis, and UCB Pharma., Grant/research support from: Vedat Hamuryudan has received grant/research support from Celgene., Izzet Fresko: None declared, Melike Melikoglu: None declared, Emire Seyahi Speakers bureau: Emire Seyahi has received honorariums for presentations from Novartis, Pfizer, AbbVie, and Gliead., Serdal Ugurlu: None declared, Huri Ozdogan: None declared, Hasan Yazici: None declared, Gulen Hatemi Speakers bureau: Gulen Hatemi has served as a speaker for AbbVie, Celgene, Novartis, and UCB Pharma, Grant/research support from: Gulen Hatemi has received grant/research support from Celgene.

Creating learning health systems and the emerging role of biomedical informatics

INTRODUCTION

The nature of information used in medicine has changed. In the past, we were limited to routine clinical data and published clinical trials. Today, we deal with massive, multiple data streams and easy access to new tests, ideas, and capabilities to process them. Whereas in the past getting information for decision-making was a challenge, now, it is how to analyze, evaluate and prioritize all that is readily available through the multitude of data-collecting devices. Clinicians must become adept with the tools needed to deal with the era of big data, requiring a major change in how we learn to make decisions. Major change is often met with resistance and questions about value. A Learning Health System is an enabler to encourage the development of such tools and demonstrate value in improved decision-making.

METHODS

We describe how we are developing a Biomedical Informatics program to help our medical institution's evolution as an academic Learning Health System, including strategy, training for house staff and examples of the role of informatics from operations to research.

RESULTS

We described an array of learning health system implementations and educational programs to improve healthcare and prepare a cadre of physicians with basic information technology skills. The programs have been well accepted with, for example, increasing interest and enrollment in the educational programs.

CONCLUSIONS

We are now in an era when large volumes of a wide variety of data are readily available. The challenge is not so much in the acquisition of data, but in assessing the quality, relevance and value of the data. The data we can get may not be the data we need. In the past, sources of data were limited, and trial results published in journals were the major source of evidence for decision making. The advent of powerful analytics systems has changed the concept of evidence. Clinicians will have to develop the skills necessary to work in the era of big data. It is not reasonable to expect that all clinicians will also be data scientists. However, understanding the role of AI and predictive analytics, and how to apply them, will become progressively more important. Programs such as the one being implemented at Wake Forest fill that need.

CXCL1: A new diagnostic biomarker for human tuberculosis discovered using Diversity Outbred mice

More humans have died of tuberculosis (TB) than any other infectious disease and millions still die each year. Experts advocate for blood-based, serum protein biomarkers to help diagnose TB, which afflicts millions of people in high-burden countries. However, the protein biomarker pipeline is small. Here, we used the Diversity Outbred (DO) mouse population to address this gap, identifying five protein biomarker candidates. One protein biomarker, serum CXCL1, met the World Health Organization’s Targeted Product Profile for a triage test to diagnose active TB from latent M.tb infection (LTBI), non-TB lung disease, and normal sera in HIV-negative, adults from South Africa and Vietnam. To find the biomarker candidates, we quantified seven immune cytokines and four inflammatory proteins corresponding to highly expressed genes unique to progressor DO mice. Next, we applied statistical and machine learning methods to the data, i.e., 11 proteins in lungs from 453 infected and 29 non-infected mice. After searching all combinations of five algorithms and 239 protein subsets, validating, and testing the findings on independent data, two combinations accurately diagnosed progressor DO mice: Logistic Regression using MMP8; and Gradient Tree Boosting using a panel of 4: CXCL1, CXCL2, TNF, IL-10. Of those five protein biomarker candidates, two (MMP8 and CXCL1) were crucial for classifying DO mice; were above the limit of detection in most human serum samples; and had not been widely assessed for diagnostic performance in humans before. In patient sera, CXCL1 exceeded the triage diagnostic test criteria (>90% sensitivity; >70% specificity), while MMP8 did not. Using Area Under the Curve analyses, CXCL1 averaged 94.5% sensitivity and 88.8% specificity for active pulmonary TB (ATB) vs LTBI; 90.9% sensitivity and 71.4% specificity for ATB vs non-TB; and 100.0% sensitivity and 98.4% specificity for ATB vs normal sera. Our findings overall show that the DO mouse population can discover diagnostic-quality, serum protein biomarkers of human TB.

More humans die of tuberculosis (TB) than any other infectious disease, yet diagnostic tools remain limited. Here, we used the Diversity Outbred mouse population to discover candidate protein biomarkers of human TB. By applying statistical methods and machine learning to multidimensional data, we identified CXCL1 and MMP8 as the two most promising protein biomarker candidates. When evaluated in samples from human patients, CXCL1 achieved the World Health Organization’s targeted profile for a triage diagnostic test, discriminating active TB from important clinical differential diagnoses: latent Mtb infection and non-TB lung disease in HIV-negative adults. Overall, our studies show how a translationally relevant animal population model can accelerate TB biomarker discovery, validation, and testing for humans.

TMOD-35. PREDICTION OF OVERALL SURVIVAL, AND MOLECULAR MARKERS IN GLIOMAS VIA ANALYSIS OF DIGITAL PATHOLOGY IMAGES USING DEEP LEARNING

BACKGROUND

Microscopic features of brain tumors, such as tumor cell morphology, type/degree of microvascular hyperplasia, mitotic activity, and extent of zonal/geographic necrosis, among many others, are measurable and reflect underlying molecular markers that are predictive of patient prognosis, signifying that quantitative analysis may provide insight into disease mechanics. We developed a computational method to predict overall-survival and molecular markers for brain tumors using deep learning on whole-slide digital images (WSDI).

METHODS

The WSDI were acquired from TCGA for 663 patients [IDH:333 wildtype, 330 mutants, 1p/19q:201 non-codeleted, 129 codeleted]. A set of 100 region-of-interest each comprising 1024x1024 that contained viable tumor with descriptive histologic characteristics and that were free of artifacts were extracted. A modified version of ResNet with architecture of 50 convolutional layers was used. The network was optimized using stochastic gradient decent optimization method with binary cross-entropy loss. Sigmoid- and linear-activation were, respectively, used as final layer for mutation and survival prediction. Data was divided into training (50%), testing (25%), and validation (25%).

RESULTS

The model predicted IDH and 1p/19q with an accuracy of 88.92%[sensitivity(se)/ specificity(sp)=87.77/84.35] and 88.23%(se/sp=87.38/88.58), respectively. The accuracy was further improved, when classification was done within homogeneous grades, for IDH [II=90.50%(se/sp=91.52/78.57), III=91.21%(se/sp=91.89/89.47), IV=92.77%(se/sp=77.78/93.51)] and 1p/19q [II=91.51%(se/sp=91.30/91.66), III=92.56(se/sp =93.33,92.04)]. The Pearson correlation coefficient between the predicted scores and overall-survival was 0.79 (p< 0.0001).

CONCLUSION

Our findings suggest that deep learning techniques can be applied to WSDI for objective, and accurate prediction of mutations and survival. Our approach, when compared with expensive molecular based assays that invariably capture molecular markers from a small part of the tumor and also destroy the tissue, could (i) offer the same service at a reduced price, (ii) enable disease characterization across the entire landscape of the tissue, (iii) be beneficial for tissues inadequate for molecular testing, and (iv) does not need physical shipping of the tissue.

Patients' experiences with Behçet's syndrome: structured interviews among patients with different types of organ involvement

OBJECTIVES

Behçet's syndrome (BS) is a multisystem variable-vessel vasculitis with significant life impact. The aim of this study was to explore the perspectives of patients with BS with different types of organ involvement.

METHODS

Semi-structured qualitative interviews were conducted with 20 patients with BS with different types of organ involvement. Interviews were audio-recorded, transcribed, and translated into English. A Grounded Theory approach was employed in thematic analysis of translated interviews.

RESULTS

Interviews with participants yielded four themes, including symptoms (skin problems, pain, vision problems, fatigue/sleep disturbances, and gastrointestinal/weight loss), impact on function (impact on speech and vision, mobility, energy for tasks, adaptations, and self-care), psychological impact (emotions and emotional management techniques), and social impact (ability to socialize generally and impact on familial relationships).

CONCLUSIONS

Patients with BS identified several domains, including physical functioning, psychological state, and social identity that are significantly modulated by the symptoms of BS. Those are inter-related with physical symptoms, reflecting the multi-system character of BS, and impair patients' function impacting on psychological and social identities. This work advances an understanding of BS, and will be useful in developing patient-oriented outcome measures for use in studying BS.

Informatics Approaches to Address New Challenges in the Classification of Lymphoid Malignancies

Purpose

Lymphoid malignancies are remarkably heterogeneous, with variations in outcomes and clinical, biologic, and histologic presentation complicating classification according to the World Health Organization guidelines. Incorrect classification of lymphoid neoplasms can result in suboptimal therapeutic strategies for individual patients and confound the interpretation of clinical trials involving personalized, class-based treatments. This review discusses the potential role of pathology informatics in improving the classification accuracy and objectivity for lymphoid malignancies.

Design

We identified peer-reviewed publications examining pathology informatics approaches for the classification of lymphoid malignancies, reviewed developments in the lymphoma classification systems, and summarized computational methods for pathologic assessment that can impact practice.

Results

Computer-assisted pathology image analysis algorithms in lymphoma most commonly have been applied to follicular lymphoma to address biologic heterogeneity and subjectivity in the process of classification.

Conclusion

Objective methods are available to assist pathologists in lymphoma classification and grading, and have been demonstrated to provide measurable benefits in specific contexts. Future validation and extension of these approaches will require datasets that link high resolution pathology images available for image analysis algorithms with clinical variables and follow up outcomes.

Computer-assisted quantification of CD3+ T cells in follicular lymphoma

The advance of high resolution digital scans of pathology slides allowed development of computer based image analysis algorithms that may help pathologists in IHC stains quantification. While very promising, these methods require further refinement before they are implemented in routine clinical setting. Particularly critical is to evaluate algorithm performance in a setting similar to current clinical practice. In this article, we present a pilot study that evaluates the use of a computerized cell quantification method in the clinical estimation of CD3 positive (CD3+) T cells in follicular lymphoma (FL). Our goal is to demonstrate the degree to which computerized quantification is comparable to the practice of estimation by a panel of expert pathologists. The computerized quantification method uses entropy based histogram thresholding to separate brown (CD3+) and blue (CD3-) regions after a color space transformation. A panel of four board-certified hematopathologists evaluated a database of 20 FL images using two different reading methods: visual estimation and manual marking of each CD3+ cell in the images. These image data and the readings provided a reference standard and the range of variability among readers. Sensitivity and specificity measures of the computer's segmentation of CD3+ and CD- T cell are recorded. For all four pathologists, mean sensitivity and specificity measures are 90.97 and 88.38%, respectively. The computerized quantification method agrees more with the manual cell marking as compared to the visual estimations. Statistical comparison between the computerized quantification method and the pathologist readings demonstrated good agreement with correlation coefficient values of 0.81 and 0.96 in terms of Lin's concordance correlation and Spearman's correlation coefficient, respectively. These values are higher than most of those calculated among the pathologists. In the future, the computerized quantification method may be used to investigate the relationship between the overall architectural pattern (i.e., interfollicular vs. follicular) and outcome measures (e.g., overall survival, and time to treatment). © 2017 International Society for Advancement of Cytometry.

APPLYING HUMAN FACTORS ENGINEERING TO IMPROVE USABILITY AND WORKFLOW IN PATHOLOGY INFORMATICS

Human factors engineering is an underutilized approach in the design, evaluation, and implementation of health information technology. Heuristic evaluation of the usability of an interface is a 'low-hanging fruit' for identifying a set of relatively simple modifications to a software program that can make software easier to use. In this paper, we describe recommendations to improve the usability of a software package used to view digitized images of tissues by pathologists. Several recommendations were immediately implemented, and others are planned for future releases. The changes are anticipated to be more compatible with user expectations from interacting with similar elements in other packages, and thus easier to learn and use.

Digital otoscopy versus microscopy: How correct and confident are ear experts in their diagnoses?

Introduction With the growing popularity of telemedicine and tele-diagnostics, clinical validation of new devices is essential. This study sought to investigate whether high-definition digital still images of the eardrum provide sufficient information to make a correct diagnosis, as compared with the gold standard view provided by clinical microscopy. Methods Twelve fellowship-trained ear physicians (neurotologists) reviewed the same set of 210 digital otoscope eardrum images. Participants diagnosed each image as normal or, if abnormal, they selected from seven types of ear pathology. Diagnostic percentage correct for each pathology was compared with a gold standard of diagnosis using clinical microscopy with adjunct audiometry and/or tympanometry. Participants also rated their degree of confidence for each diagnosis. Results Overall correctness of diagnosis for ear pathologies ranged from 48.6-100%, depending on the type of pathology. Neurotologists were 72% correct in identifying eardrums as normal. Reviewers' confidence in diagnosis varied substantially among types of pathology, as well as among participants. Discussion High-definition digital still images of eardrums provided sufficient information for neurotologists to make correct diagnoses for some pathologies. However, some diagnoses, such as middle ear effusion, were more difficult to diagnose when based only on a still image. Levels of confidence of reviewers did not generally correlate with diagnostic ability.

Automated fluorescent miscroscopic image analysis of PTBP1 expression in glioma

Multiplexed immunofluorescent testing has not entered into diagnostic neuropathology due to the presence of several technical barriers, amongst which includes autofluorescence. This study presents the implementation of a methodology capable of overcoming the visual challenges of fluorescent microscopy for diagnostic neuropathology by using automated digital image analysis, with long term goal of providing unbiased quantitative analyses of multiplexed biomarkers for solid tissue neuropathology. In this study, we validated PTBP1, a putative biomarker for glioma, and tested the extent to which immunofluorescent microscopy combined with automated and unbiased image analysis would permit the utility of PTBP1 as a biomarker to distinguish diagnostically challenging surgical biopsies. As a paradigm, we utilized second resections from patients diagnosed either with reactive brain changes (pseudoprogression) and recurrent glioblastoma (true progression). Our image analysis workflow was capable of removing background autofluorescence and permitted quantification of DAPI-PTBP1 positive cells. PTBP1-positive nuclei, and the mean intensity value of PTBP1 signal in cells. Traditional pathological interpretation was unable to distinguish between groups due to unacceptably high discordance rates amongst expert neuropathologists. Our data demonstrated that recurrent glioblastoma showed more DAPI-PTBP1 positive cells and a higher mean intensity value of PTBP1 signal compared to resections from second surgeries that showed only reactive gliosis. Our work demonstrates the potential of utilizing automated image analysis to overcome the challenges of implementing fluorescent microscopy in diagnostic neuropathology.

Visually Meaningful Histopathological Features for Automatic Grading of Prostate Cancer

Histopathologic features, particularly Gleason grading system, have contributed significantly to the diagnosis, treatment, and prognosis of prostate cancer for decades. However, prostate cancer demonstrates enormous heterogeneity in biological behavior, thus establishing improved prognostic and predictive markers is particularly important to personalize therapy of men with clinically localized and newly diagnosed malignancy. Many automated grading systems have been developed for Gleason grading but acceptance in the medical community has been lacking due to poor interpretability. To overcome this problem, we developed a set of visually meaningful features to differentiate between low- and high-grade prostate cancer. The visually meaningful feature set consists of luminal and architectural features. For luminal features, we compute: 1) the shortest path from the nuclei to their closest luminal spaces; 2) ratio of the epithelial nuclei to the total number of nuclei. A nucleus is considered an epithelial nucleus if the shortest path between it and the luminal space does not contain any other nucleus; 3) average shortest distance of all nuclei to their closest luminal spaces. For architectural features, we compute directional changes in stroma and nuclei using directional filter banks. These features are utilized to create two subspaces; one for prostate images histopathologically assessed as low grade and the other for high grade. The grade associated with a subspace, which results in the minimum reconstruction error is considered as the prediction for the test image. For training, we utilized 43 regions of interest (ROI) images, which were extracted from 25 prostate whole slide images of The Cancer Genome Atlas (TCGA) database. For testing, we utilized an independent dataset of 88 ROIs extracted from 30 prostate whole slide images. The method resulted in 93.0% and 97.6% training and testing accuracies, respectively, for the spectrum of cases considered. The application of visually meaningful features provided promising levels of accuracy and consistency for grading prostate cancer.

Lung necrosis and neutrophils reflect common pathways of susceptibility to Mycobacterium tuberculosis in genetically diverse, immune-competent mice

Pulmonary tuberculosis (TB) is caused by Mycobacterium tuberculosis in susceptible humans. Here, we infected Diversity Outbred (DO) mice with ∼100 bacilli by aerosol to model responses in a highly heterogeneous population. Following infection, ‘supersusceptible’, ‘susceptible’ and ‘resistant’ phenotypes emerged. TB disease (reduced survival, weight loss, high bacterial load) correlated strongly with neutrophils, neutrophil chemokines, tumor necrosis factor (TNF) and cell death. By contrast, immune cytokines were weak correlates of disease. We next applied statistical and machine learning approaches to our dataset of cytokines and chemokines from lungs and blood. Six molecules from the lung: TNF, CXCL1, CXCL2, CXCL5, interferon-γ (IFN-γ), interleukin 12 (IL-12); and two molecules from blood – IL-2 and TNF – were identified as being important by applying both statistical and machine learning methods. Using molecular features to generate tree classifiers, CXCL1, CXCL2 and CXCL5 distinguished four classes (supersusceptible, susceptible, resistant and non-infected) from each other with approximately 77% accuracy using completely independent experimental data. By contrast, models based on other molecules were less accurate. Low to no IFN-γ, IL-12, IL-2 and IL-10 successfully discriminated non-infected mice from infected mice but failed to discriminate disease status amongst supersusceptible, susceptible and resistant M.-tuberculosis-infected DO mice. Additional analyses identified CXCL1 as a promising peripheral biomarker of disease and of CXCL1 production in the lungs. From these results, we conclude that: (1) DO mice respond variably to M. tuberculosis infection and will be useful to identify pathways involving necrosis and neutrophils; (2) data from DO mice is suited for machine learning methods to build, validate and test models with independent data based solely on molecular biomarkers; (3) low levels of immunological cytokines best indicate a lack of exposure to M. tuberculosis but cannot distinguish infection from disease.

Summary: Molecular biomarkers of tuberculosis are identified and used to classify disease status of Diversity Outbred mice that have been infected with Mycobacterium tuberculosis.

Biomedical imaging ontologies: A survey and proposal for future work

Background:

Ontology is one strategy for promoting interoperability of heterogeneous data through consistent tagging. An ontology is a controlled structured vocabulary consisting of general terms (such as “cell” or “image” or “tissue” or “microscope”) that form the basis for such tagging. These terms are designed to represent the types of entities in the domain of reality that the ontology has been devised to capture; the terms are provided with logical definitions thereby also supporting reasoning over the tagged data.

Aim:

This paper provides a survey of the biomedical imaging ontologies that have been developed thus far. It outlines the challenges, particularly faced by ontologies in the fields of histopathological imaging and image analysis, and suggests a strategy for addressing these challenges in the example domain of quantitative histopathology imaging.

Results and Conclusions:

The ultimate goal is to support the multiscale understanding of disease that comes from using interoperable ontologies to integrate imaging data with clinical and genomics data.

Impact of diffusion barriers to small cytotoxic molecules on the efficacy of immunotherapy in breast cancer

Molecular-focused cancer therapies, e.g., molecularly targeted therapy and immunotherapy, so far demonstrate only limited efficacy in cancer patients. We hypothesize that underestimating the role of biophysical factors that impact the delivery of drugs or cytotoxic cells to the target sites (for associated preferential cytotoxicity or cell signaling modulation) may be responsible for the poor clinical outcome. Therefore, instead of focusing exclusively on the investigation of molecular mechanisms in cancer cells, convection-diffusion of cytotoxic molecules and migration of cancer-killing cells within tumor tissue should be taken into account to improve therapeutic effectiveness. To test this hypothesis, we have developed a mathematical model of the interstitial diffusion and uptake of small cytotoxic molecules secreted by T-cells, which is capable of predicting breast cancer growth inhibition as measured both in vitro and in vivo. Our analysis shows that diffusion barriers of cytotoxic molecules conspire with γδ T-cell scarcity in tissue to limit the inhibitory effects of γδ T-cells on cancer cells. This may increase the necessary ratios of γδ T-cells to cancer cells within tissue to unrealistic values for having an intended therapeutic effect, and decrease the effectiveness of the immunotherapeutic treatment.

Partitioning histopathological images: an integrated framework for supervised color-texture segmentation and cell splitting

For quantitative analysis of histopathological images, such as the lymphoma grading systems, quantification of features is usually carried out on single cells before categorizing them by classification algorithms. To this end, we propose an integrated framework consisting of a novel supervised cell-image segmentation algorithm and a new touching-cell splitting method. For the segmentation part, we segment the cell regions from the other areas by classifying the image pixels into either cell or extra-cellular category. Instead of using pixel color intensities, the color-texture extracted at the local neighborhood of each pixel is utilized as the input to our classification algorithm. The color-texture at each pixel is extracted by local Fourier transform (LFT) from a new color space, the most discriminant color space (MDC). The MDC color space is optimized to be a linear combination of the original RGB color space so that the extracted LFT texture features in the MDC color space can achieve most discrimination in terms of classification (segmentation) performance. To speed up the texture feature extraction process, we develop an efficient LFT extraction algorithm based on image shifting and image integral. For the splitting part, given a connected component of the segmentation map, we initially differentiate whether it is a touching-cell clump or a single nontouching cell. The differentiation is mainly based on the distance between the most likely radial-symmetry center and the geometrical center of the connected component. The boundaries of touching-cell clumps are smoothed out by Fourier shape descriptor before carrying out an iterative, concave-point and radial-symmetry based splitting algorithm. To test the validity, effectiveness and efficiency of the framework, it is applied to follicular lymphoma pathological images, which exhibit complex background and extracellular texture with nonuniform illumination condition. For comparison purposes, the results of the proposed segmentation algorithm are evaluated against the outputs of superpixel, graph-cut, mean-shift, and two state-of-the-art pathological image segmentation methods using ground-truth that was established by manual segmentation of cells in the original images. Our segmentation algorithm achieves better results than the other compared methods. The results of splitting are evaluated in terms of under-splitting, over-splitting, and encroachment errors. By summing up the three types of errors, we achieve a total error rate of 5.25% per image.

Computer-aided classification of centroblast cells in follicular lymphoma

OBJECTIVE

To distinguish centroblast cells from non-centroblast cells using a novel automated method in follicular lymphoma cases and measure its performance on cases obtained by a consensus of six pathologists.

STUDY DESIGN

Geometric and color texture features were used in the training and testing of the supervised quadratic discriminant analysis classifier. The technique was trained and tested on a data set composed of 218 centroblast images and 218 non-centroblast images. Computer performance was tested by measuring sensitivity and specificity among cells classified as centroblast and non-centroblast by consensus of six board-certified hematopathologists.

RESULTS

Automated classification distinguished centroblast from non-centroblast cells with a classification accuracy of 82.56% and sensitivity and specificity of 86.67% and 86.96%, respectively, when the approach was tested.

CONCLUSION

The novelty of our approach is the identification of the centroblast cells with prior information and the introduction of the principal component analysis in the spectral domain to extract texture color features.

New targets of therapy in T-cell lymphomas

T-cell lymphomas (TCL) are characterized by poor response to chemotherapy and generally poor outcome. While molecular profiling has identified distinct biological subsets and therapeutic targets in B-cell lymphomas, the molecular characterization of TCL has been slower. Surface markers expressed on malignant T-cells, such as CD2, CD3, CD4, CD25, and CD52 were the first TCL-specific therapeutic targets to be discovered. However, the presence of these receptors on normal T-cells means that monoclonal antibody (mAb)- or immunotoxin (IT)-based therapy in TCL inevitably results in variable degrees of immunosuppression. Thus, although some mAbs/IT have significant activity in selected subsets of TCL, more specific agents that target signaling pathways preferentially activated in malignant T-cells are needed. One such novel class of agents is represented by the histone deacetylase (HDAC) inhibitors. These molecules selectively induce apoptosis in a variety of transformed cells, including malignant T-cells, both in vitro and in vivo. Several HDAC inhibitors have been studied in TCL with promising results, and have recently been approved for clinical use. Immunomodulatory drugs, such as interferons and Toll Receptor (TLR) agonists have significant clinical activity in TCL, and are particularly important in the treatment of primary cutaneous subtypes (CTCL). Although most classical cytotoxic drugs have limited efficacy against TCL, agents that inhibit purine and pyrimidine metabolism, known as nucleoside analogues, and novel antifolate drugs, such as pralatrexate, are highly active in TCL. With improved molecular profiling of TCL novel pharmacological agents with activity in TCL are now being discovered at an increasingly rapid pace. Clinical trials are in progress and these agents are being integrated in combination therapies for TCL, both in the relapsed/refractory setting as well as front line.

Temporal analysis of tumor heterogeneity and volume for cervical cancer treatment outcome prediction: preliminary evaluation

In this paper, we present a method of quantifying the heterogeneity of cervical cancer tumors for use in radiation treatment outcome prediction. Features based on the distribution of masked wavelet decomposition coefficients in the tumor region of interest (ROI) of temporal dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) studies were used along with the imaged tumor volume to assess the response of the tumors to treatment. The wavelet decomposition combined with ROI masking was used to extract local intensity variations in the tumor. The developed method was tested on a data set consisting of 23 patients with advanced cervical cancer who underwent radiation therapy; 18 of these patients had local control of the tumor, and five had local recurrence. Each patient participated in two DCE-MRI studies: one prior to treatment and another early into treatment (2-4 weeks). An outcome of local control or local recurrence of the tumor was assigned to each patient based on a posttherapy follow-up at least 2 years after the end of treatment. Three different supervised classifiers were trained on combinational subsets of the full wavelet and volume feature set. The best-performing linear discriminant analysis (LDA) and support vector machine (SVM) classifiers each had mean prediction accuracies of 95.7%, with the LDA classifier being more sensitive (100% vs. 80%) and the SVM classifier being more specific (100% vs. 94.4%) in those cases. The K-nearest neighbor classifier performed the best out of all three classifiers, having multiple feature sets that were used to achieve 100% prediction accuracy. The use of distribution measures of the masked wavelet coefficients as features resulted in much better predictive performance than those of previous approaches based on tumor intensity values and their distributions or tumor volume alone.

Extraction of color features in the spectral domain to recognize centroblasts in histopathology

In this paper, we are proposing a novel automated method to recognize centroblast (CB) cells from non-centroblast (non-CB) cells for computer-assisted evaluation of follicular lymphoma tissue samples. The method is based on training and testing of a quadratic discriminant analysis (QDA) classifier. The novel aspects of this method are the identification of the CB object with prior information, and the introduction of the principal component analysis (PCA) in the spectral domain to extract color texture features. Both geometric and texture features are used to achieve the classification. Experimental results on real follicular lymphoma images demonstrate that the combined feature space improved the performance of the system significantly. The implemented method can identify centroblast cells (CB) from non-centroblast cells (non-CB) with a classification accuracy of 82.56%.

Feature-based registration of histopathology images with different stains: an application for computerized follicular lymphoma prognosis

Follicular lymphoma (FL) is the second most common type of non-Hodgkin's lymphoma. Manual histological grading of FL is subject to remarkable inter- and intra-reader variations. A promising approach to grading is the development of a computer-assisted system that improves consistency and precision. Correlating information from adjacent slides with different stain types requires establishing spatial correspondences between the digitized section pair through a precise non-rigid image registration. However, the dissimilar appearances of the different stain types challenges existing registration methods. This study proposes a method for the automatic non-rigid registration of histological section images with different stain types. This method is based on matching high level features that are representative of small anatomical structures. This choice of feature provides a rich matching environment, but also results in a high mismatch probability. Matching confidence is increased by establishing local groups of coherent features through geometric reasoning. The proposed method is validated on a set of FL images representing different disease stages. Statistical analysis demonstrates that given a proper feature set the accuracy of automatic registration is comparable to manual registration.

VirtualPACS--a federating gateway to access remote image data resources over the grid

Collaborations in biomedical research and clinical studies require that data, software, and computational resources be shared between geographically distant institutions. In radiology, there is a related issue of sharing remote DICOM data over the Internet. This paper focuses on the problem of federating multiple image data resources such that clients can interact with them as if they are stored in a centralized PACS. We present a toolkit, called VirtualPACS, to support this functionality. Using the toolkit, users can perform standard DICOM operations (query, retrieve, and submit) across distributed image databases. The key features of the toolkit are: (1) VirtualPACS makes it easy to use existing DICOM client applications for data access; (2) it can easily be incorporated into an imaging workflow as a DICOM source; (3) using VirtualPACS, heterogeneous collections of DICOM sources are exposed to clients through a uniform interface and common data model; and (4) DICOM image databases without DICOM messaging can be accessed.

e-Science, caGrid, and Translational Biomedical Research

Translational research projects target a wide variety of diseases, test many different kinds of biomedical hypotheses, and employ a large assortment of experimental methodologies. Diverse data, complex execution environments, and demanding security and reliability requirements make the implementation of these projects extremely challenging and require novel e-Science technologies.

Adaptive discriminant wavelet packet transform and local binary patterns for meningioma subtype classification

The inherent complexity and non-homogeneity of texture makes classification in medical image analysis a challenging task. In this paper, we propose a combined approach for meningioma subtype classification using subband texture (macro) features and micro-texture features. These are captured using the Adaptive Wavelet Packet Transform (ADWPT) and Local Binary Patterns (LBPs), respectively. These two different textural features are combined together and used for classification. The effect of various dimensionality reduction techniques on classification performance is also investigated. We show that high classification accuracies can be achieved using ADWPT. Although LBP features do not provide higher overall classification accuracies than ADWPT, it manages to provide higher accuracy for a meningioma subtype that is difficult to classify otherwise.

Analysis of temporal changes of mammographic features: computer-aided classification of malignant and benign breast masses

A new classification scheme was developed to classify mammographic masses as malignant and benign by using interval change information. The masses on both the current and the prior mammograms were automatically segmented using an active contour method. From each mass, 20 run length statistics (RLS) texture features, 3 speculation features, and 12 morphological features were extracted. Additionally, 20 difference RLS features were obtained by subtracting the prior RLS features from the corresponding current RLS features. The feature space consisted of the current RLS features, the difference RLS features, the current and prior speculation features, and the current and prior mass sizes. Stepwise feature selection and linear discriminant analysis classification were used to select and merge the most useful features. A leave-one-case-out resampling scheme was used to train and test the classifier using 140 temporal image pairs (85 malignant, 55 benign) obtained from 57 biopsy-proven masses (33 malignant, 24 benign) in 56 patients. An average of 10 features were selected from the 56 training subsets: 4 difference RLS features, 4 RLS features, and 1 speculation feature from the current image, and 1 speculation feature from the prior, were most often chosen. The classifier achieved an average training Az of 0.92 and a test Az of 0.88. For comparison, a classifier was trained and tested using features extracted from the 120 current single images. This classifier achieved an average training Az of 0.90 and a test Az of 0.82. The information on the prior image significantly (p = 0.015) improved the accuracy for classification of the masses.