Longitudinal studies play a key role in various fields, including epidemiology, clinical research, and genomic analysis. Currently, the most popular methods in longitudinal data analysis are model-driven regression approaches, which impose strong prior assumptions and are unable to scale to large problems in the manner of machine learning algorithms. In this work, we propose a novel longitudinal support vector regression (LSVR) algorithm that not only takes the advantage of one of the most popular machine learning methods, but also is able to model the temporal nature of longitudinal data by taking into account observational dependence within subjects. We test LSVR on publicly available data from the DREAM-Phil Bowen ALS Prediction Prize4Life challenge. Results suggest that LSVR is at a minimum competitive with favored machine learning methods and is able to outperform those methods in predicting ALS score one month in advance.

Mapping job titles to standardized occupation classification (SOC) codes is an important step in evaluating changes in health risks over time as measured in inspection databases. However, manual SOC coding is cost prohibitive for very large studies. Computer based SOC coding systems can improve the efficiency of incorporating occupational risk factors into large-scale epidemiological studies. We present a novel method of mapping verbatim job titles to SOC codes using a large table of prior knowledge available in the public domain that included detailed description of the tasks and activities and their synonyms relevant to each SOC code. Job titles are compared to our knowledge base to find the closest matching SOC code. A soft Jaccard index is used to measure the similarity between a previously unseen job title and the knowledge base. Additional information such as standardized industrial codes can be incorporated to improve the SOC code determination by providing additional context to break ties in matches.

Clinical research approval processes are complex since they involve human subject welfare as well as regulatory and ethical concerns. Typically, clinical research institutions have an elaborate established review process; the labor-intensive and time-consuming nature of this process can result in approval delays thus significantly impacting the biomedical research. This paper describes the development of a Web-based information system designed to make the process of protocol approval more efficient, organized, and accurate. The system enables the principal investigators (Pis) to initiate, track and manage clinical studies, monitor review processes, view status changes, and quickly respond to requests for additional information. In addition, it allows the protocol coordinators (PCs) to manage the entire Institutional Review Board (IRB) approval process.

This paper describes potential strategies for data modeling and implementation as part of the general architecture of CSIS, a clinical study and informatics system that has been developed for the National Institute of Neuroscience and Stroke (NINDS). We discuss the NINDS requirements and how they influenced the system design, with an emphasis on dynamic form creation. We also evaluate open standards such as "CDISC ODM" and several database technology choices, namely conventional relational "EAV" tables and the new "XMLDB" from Oracle. We describe performance test results, which show that it is feasible to implement CDISC ODM in Oracle XMLDB and the schema-based storage option for storing XML content is better than "CLOB"-based for applications that require high numbers of queries for XML fragments. For storage and retrieval of whole XML documents, CLOB storage is optimal. CSIS implements a hybrid approach of combining CLOB storage for storing XML documents with relational tables for storing metadata. Due to the complexity of the CDISC schema and related performance and implementation concerns, we have decided not to implement CDISC internal storage in our system.

Interior-point methods have been successfully applied to a wide variety of linear and nonlinear programming applications. This paper presents a class of algorithms, based on path-following interior-point methodology, for performing regularized maximum-likelihood (ML) reconstructions on three-dimensional (3-D) emission tomography data. The algorithms solve a sequence of subproblems that converge to the regularized maximum likelihood solution from the interior of the feasible region (the nonnegative orthant). We propose two methods, a primal method which updates only the primal image variables and a primal-dual method which simultaneously updates the primal variables and the Lagrange multipliers. A parallel implementation permits the interior-point methods to scale to very large reconstruction problems. Termination is based on well-defined convergence measures, namely, the Karush-Kuhn-Tucker first-order necessary conditions for optimality. We demonstrate the rapid convergence of the path-following interior-point methods using both data from a small animal scanner and Monte Carlo simulated data. The proposed methods can readily be applied to solve the regularized, weighted least squares reconstruction problem.

Rat brain images acquired with a small animal positron emission tomography (PET) camera and reconstructed with the three-dimensional (3-D) ordered-subsets expectation-maximization (OSEM) algorithm with resolution recovery have better quality when the brain is imaged by itself than when inside the head with surrounding background activity. The purpose of this study was to characterize the dependence of this effect on the level of background activity, attenuation, and scatter. Monte Carlo simulations of the imaging system were performed. The coefficient of variation from replicate images, full-width at half-maximum (FWHM) from point sources and image profile fitting, and image contrast and uniformity were used to evaluate algorithm performance. A rat head with the typical levels of five and ten times the brain activity in the surrounding background requires additional iterations to achieve the same resolution as the brain-only case at a cost of 24% and 64% additional noise, respectively. For the same phantoms, object scatter reduced contrast by 3%-5%. However, attenuation degraded resolution by 0.2 mm and was responsible for up to 12% nonuniformity in the brain images suggesting that attenuation correction is useful. Given the effects of emission and attenuation distribution on both resolution and noise, simulations or phantom studies should be used for each imaging situation to select the appropriate number of OSEM iterations to achieve the desired resolution-noise levels.

Three dimensional image reconstruction from cryoelectron micrographs allows the capsid structures of icosahedral viruses to be studied. The most computationally demanding stages of the reconstruction process are those that find and refine the orientations of the virus particles. We have devised and implemented parallel solutions to the problem of determining these orientations. By enabling determination of the orientations of far more particles than previously had been possible, parallel processing methods have contributed to improvements in the quality of recent reconstructions.>