Predicting a multi-parametric probability map of active tumor extent using random forests.

Glioblastoma Mulitforme is highly infiltrative, making precise delineation of tumor margin difficult. Multimodality or multi-parametric MR imaging sequences promise an advantage over anatomic sequences such as post contrast enhancement as methods for determining the spatial extent of tumor involvement. In considering multi-parametric imaging sequences however, manual image segmentation and classification is time-consuming and prone to error. As a preliminary step toward integration of multi-parametric imaging into clinical assessments of primary brain tumors, we propose a machine-learning based multi-parametric approach that uses radiologist generated labels to train a classifier that is able to classify tissue on a voxel-wise basis and automatically generate a tumor segmentation. A random forests classifier was trained using a leave-one-out experimental paradigm. A simple linear classifier was also trained for comparison. The random forests classifier accurately predicted radiologist generated segmentations and tumor extent.

The Analytic Information Warehouse (AIW): a platform for analytics using electronic health record data.

OBJECTIVE: To create an analytics platform for specifying and detecting clinical phenotypes and other derived variables in electronic health record (EHR) data for quality improvement investigations. MATERIALS AND METHODS: We have developed an architecture for an Analytic Information Warehouse (AIW). It supports transforming data represented in different physical schemas into a common data model, specifying derived variables in terms of the common model to enable their reuse, computing derived variables while enforcing invariants and ensuring correctness and consistency of data transformations, long-term curation of derived data, and export of derived data into standard analysis tools. It includes software that implements these features and a computing environment that enables secure high-performance access to and processing of large datasets extracted from EHRs. RESULTS: We have implemented and deployed the architecture in production locally. The software is available as open source. We have used it as part of hospital operations in a project to reduce rates of hospital readmission within 30days. The project examined the association of over 100 derived variables representing disease and co-morbidity phenotypes with readmissions in 5years of data from our institution's clinical data warehouse and the UHC Clinical Database (CDB). The CDB contains administrative data from over 200 hospitals that are in academic medical centers or affiliated with such centers. DISCUSSION AND CONCLUSION: A widely available platform for managing and detecting phenotypes in EHR data could accelerate the use of such data in quality improvement and comparative effectiveness studies.

Prediction of heart rate response to conclusion of the spontaneous breathing trial by fluctuation dissipation theory.

The non-equilibrium fluctuation dissipation theorem is applied to predict how critically ill patients respond to treatment, based upon data currently collected by standard hospital monitoring devices. This framework is demonstrated on a common procedure in critical care: the spontaneous breathing trial. It is shown that the responses of groups of similar patients to the spontaneous breathing trial can be predicted by the non-equilibrium fluctuation dissipation approach. This mathematical framework, when fully formed and applied to other clinical interventions, may serve as part of the basis for personalized critical care.

Automated motif discovery from glycan array data.

Assessing interactions of a glycan-binding protein (GBP) or lectin with glycans on a microarray generates large datasets, making it difficult to identify a glycan structural motif or determinant associated with the highest apparent binding strength of the GBP. We have developed a computational method, termed GlycanMotifMiner, that uses the relative binding of a GBP with glycans within a glycan microarray to automatically reveal the glycan structural motifs recognized by a GBP. We implemented the software with a web-based graphical interface for users to explore and visualize the discovered motifs. The utility of GlycanMotifMiner was determined using five plant lectins, SNA, HPA, PNA, Con A, and UEA-I. Data from the analyses of the lectins at different protein concentrations were processed to rank the glycans based on their relative binding strengths. The motifs, defined as glycan substructures that exist in a large number of the bound glycans and few non-bound glycans, were then discovered by our algorithm and displayed in a web-based graphical user interface ( http://glycanmotifminer.emory.edu ). The information is used in defining the glycan-binding specificity of GBPs. The results were compared to the known glycan specificities of these lectins generated by manual methods. A more complex analysis was also carried out using glycan microarray data obtained for a recombinant form of human galectin-8. Results for all of these lectins show that GlycanMotifMiner identified the major motifs known in the literature along with some unexpected novel binding motifs.

Integrated morphologic analysis for the identification and characterization of disease subtypes.

BACKGROUND AND OBJECTIVE: Morphologic variations of disease are often linked to underlying molecular events and patient outcome, suggesting that quantitative morphometric analysis may provide further insight into disease mechanisms. In this paper a methodology for the subclassification of disease is developed using image analysis techniques. Morphologic signatures that represent patient-specific tumor morphology are derived from the analysis of hundreds of millions of cells in digitized whole slide images. Clustering these signatures aggregates tumors into groups with cohesive morphologic characteristics. This methodology is demonstrated with an analysis of glioblastoma, using data from The Cancer Genome Atlas to identify a prognostically significant morphology-driven subclassification, in which clusters are correlated with transcriptional, genetic, and epigenetic events. MATERIALS AND METHODS: Methodology was applied to 162 glioblastomas from The Cancer Genome Atlas to identify morphology-driven clusters and their clinical and molecular correlates. Signatures of patient-specific tumor morphology were generated from analysis of 200 million cells in 462 whole slide images. Morphology-driven clusters were interrogated for associations with patient outcome, response to therapy, molecular classifications, and genetic alterations. An additional layer of deep, genome-wide analysis identified characteristic transcriptional, epigenetic, and copy number variation events. RESULTS AND DISCUSSION: Analysis of glioblastoma identified three prognostically significant patient clusters (median survival 15.3, 10.7, and 13.0 months, log rank p=1.4e-3). Clustering results were validated in a separate dataset. Clusters were characterized by molecular events in nuclear compartment signaling including developmental and cell cycle checkpoint pathways. This analysis demonstrates the potential of high-throughput morphometrics for the subclassification of disease, establishing an approach that complements genomics.

Leveraging derived data elements in data analytic models for understanding and predicting hospital readmissions.

Hospital readmissions depend on numerous factors. Automated risk calculation using electronic health record (EHR) data could allow targeting care to prevent them. EHRs usually are incomplete with respect to data relevant to readmissions prediction. Lack of standard data representations in EHRs restricts generalizability of predictive models. We propose developing such models by first generating derived variables that characterize clinical phenotype. This reduces the number of variables, reduces noise, introduces clinical knowledge into model building, and abstracts away the underlying data representation, thus facilitating use of standard data mining algorithms. We combined this pre-processing step with a random forest algorithm to compute risk for readmission within 30 days for patients in ten disease categories. Results were promising for encounters that our algorithm assigned very high or very low risk. Assigning patients to either of these two risk groups could be of value to patient care teams aiming to prevent readmissions.

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.

Pathology is a medical subspecialty that practices the diagnosis of disease. Microscopic examination of tissue reveals information enabling the pathologist to render accurate diagnoses and to guide therapy. The basic process by which anatomic pathologists render diagnoses has remained relatively unchanged over the last century, yet advances in information technology now offer significant opportunities in image-based diagnostic and research applications. Pathology has lagged behind other healthcare practices such as radiology where digital adoption is widespread. As devices that generate whole slide images become more practical and affordable, practices will increasingly adopt this technology and eventually produce an explosion of data that will quickly eclipse the already vast quantities of radiology imaging data. These advances are accompanied by significant challenges for data management and storage, but they also introduce new opportunities to improve patient care by streamlining and standardizing diagnostic approaches and uncovering disease mechanisms. Computer-based image analysis is already available in commercial diagnostic systems, but further advances in image analysis algorithms are warranted in order to fully realize the benefits of digital pathology in medical discovery and patient care. In coming decades, pathology image analysis will extend beyond the streamlining of diagnostic workflows and minimizing interobserver variability and will begin to provide diagnostic assistance, identify therapeutic targets, and predict patient outcomes and therapeutic responses.

The proneural molecular signature is enriched in oligodendrogliomas and predicts improved survival among diffuse gliomas.

The Cancer Genome Atlas Project (TCGA) has produced an extensive collection of '-omic' data on glioblastoma (GBM), resulting in several key insights on expression signatures. Despite the richness of TCGA GBM data, the absence of lower grade gliomas in this data set prevents analysis genes related to progression and the uncovering of predictive signatures. A complementary dataset exists in the form of the NCI Repository for Molecular Brain Neoplasia Data (Rembrandt), which contains molecular and clinical data for diffuse gliomas across the full spectrum of histologic class and grade. Here we present an investigation of the significance of the TCGA consortium's expression classification when applied to Rembrandt gliomas. We demonstrate that the proneural signature predicts improved clinical outcome among 176 Rembrandt gliomas that includes all histologies and grades, including GBMs (log rank test p = 1.16e-6), but also among 75 grade II and grade III samples (p  =  2.65e-4). This gene expression signature was enriched in tumors with oligodendroglioma histology and also predicted improved survival in this tumor type (n =  43, p  =  1.25e-4). Thus, expression signatures identified in the TCGA analysis of GBMs also have intrinsic prognostic value for lower grade oligodendrogliomas, and likely represent important differences in tumor biology with implications for treatment and therapy. Integrated DNA and RNA analysis of low-grade and high-grade proneural gliomas identified increased expression and gene amplification of several genes including GLIS3, TGFB2, TNC, AURKA, and VEGFA in proneural GBMs, with corresponding loss of DLL3 and HEY2. Pathway analysis highlights the importance of the Notch and Hedgehog pathways in the proneural subtype. This demonstrates that the expression signatures identified in the TCGA analysis of GBMs also have intrinsic prognostic value for low-grade oligodendrogliomas, and likely represent important differences in tumor biology with implications for treatment and therapy.

An integrative approach for in silico glioma research.

The integration of imaging and genomic data is critical to forming a better understanding of disease. Large public datasets, such as The Cancer Genome Atlas, present a unique opportunity to integrate these complementary data types for in silico scientific research. In this letter, we focus on the aspect of pathology image analysis and illustrate the challenges associated with analyzing and integrating large-scale image datasets with molecular characterizations. We present an example study of diffuse glioma brain tumors, where the morphometric analysis of 81 million nuclei is integrated with clinically relevant transcriptomic and genomic characterizations of glioblastoma tumors. The preliminary results demonstrate the potential of combining morphometric and molecular characterizations for in silico research.

Localized content-based image retrieval.

We define localized content-based image retrieval as a CBIR task where the user is only interested in a portion of the image, and the rest of the image is irrelevant. In this paper we present a localized CBIR system, Accio, that uses labeled images in conjunction with a multiple-instance learning algorithm to first identify the desired object and weight the features accordingly, and then to rank images in the database using a similarity measure that is based upon only the relevant portions of the image. A challenge for localized CBIR is how to represent the image to capture the content. We present and compare two novel image representations, which extend traditional segmentation-based and salient point-based techniques respectively, to capture content in a localized CBIR setting.