Development of automated image analysis tools for verification of radiotherapy field accuracy with an electronic portal imaging device

The successful management of cancer with radiation relies on the accurate deposition of a prescribed dose to a prescribed anatomical volume within the patient. Treatment set-up errors are inevitable because the alignment of field shaping devices with the patient must be repeated daily up to eighty times during the course of a fractionated radiotherapy treatment. With the invention of electronic portal imaging devices (EPIDs), patient's portal images can be visualized daily in real-time after only a small fraction of the radiation dose has been delivered to each treatment field. However, the accuracy of human visual evaluation of low-contrast portal images has been found to be inadequate. The goal of this research is to develop automated image analysis tools to detect both treatment field shape errors and patient anatomy placement errors with an EPID. A moments method has been developed to align treatment field images to compensate for lack of repositioning precision of the image detector. A figure of merit has also been established to verify the shape and rotation of the treatment fields. Following proper alignment of treatment field boundaries, a cross-correlation method has been developed to detect shifts of the patient's anatomy relative to the treatment field boundary. Phantom studies showed that the moments method aligned the radiation fields to within 0.5mm of translation and 0.5$\sp\circ$ of rotation and that the cross-correlation method aligned anatomical structures inside the radiation field to within 1 mm of translation and 1$\sp\circ$ of rotation. A new procedure of generating and using digitally reconstructed radiographs (DRRs) at megavoltage energies as reference images was also investigated. The procedure allowed a direct comparison between a designed treatment portal and the actual patient setup positions detected by an EPID. Phantom studies confirmed the feasibility of the methodology. Both the moments method and the cross-correlation technique were implemented within an experimental radiotherapy picture archival and communication system (RT-PACS) and were used clinically to evaluate the setup variability of two groups of cancer patients treated with and without an alpha-cradle immobilization aid. The tools developed in this project have proven to be very effective and have played an important role in detecting patient alignment errors and field-shape errors in treatment fields formed by a multileaf collimator (MLC). ^

NEW TOOLS FOR MONITORING GAMMA CAMERA UNIFORMITY

Detector uniformity is a fundamental performance characteristic of all modern gamma camera systems, and ensuring a stable, uniform detector response is critical for maintaining clinical images that are free of artifact. For these reasons, the assessment of detector uniformity is one of the most common activities associated with a successful clinical quality assurance program in gamma camera imaging. The evaluation of this parameter, however, is often unclear because it is highly dependent upon acquisition conditions, reviewer expertise, and the application of somewhat arbitrary limits that do not characterize the spatial location of the non-uniformities. Furthermore, as the goal of any robust quality control program is the determination of significant deviations from standard or baseline conditions, clinicians and vendors often neglect the temporal nature of detector degradation (1). This thesis describes the development and testing of new methods for monitoring detector uniformity. These techniques provide more quantitative, sensitive, and specific feedback to the reviewer so that he or she may be better equipped to identify performance degradation prior to its manifestation in clinical images. The methods exploit the temporal nature of detector degradation and spatially segment distinct regions-of-non-uniformity using multi-resolution decomposition. These techniques were tested on synthetic phantom data using different degradation functions, as well as on experimentally acquired time series floods with induced, progressively worsening defects present within the field-of-view. The sensitivity of conventional, global figures-of-merit for detecting changes in uniformity was evaluated and compared to these new image-space techniques. The image-space algorithms provide a reproducible means of detecting regions-of-non-uniformity prior to any single flood image’s having a NEMA uniformity value in excess of 5%. The sensitivity of these image-space algorithms was found to depend on the size and magnitude of the non-uniformities, as well as on the nature of the cause of the non-uniform region. A trend analysis of the conventional figures-of-merit demonstrated their sensitivity to shifts in detector uniformity. The image-space algorithms are computationally efficient. Therefore, the image-space algorithms should be used concomitantly with the trending of the global figures-of-merit in order to provide the reviewer with a richer assessment of gamma camera detector uniformity characteristics.

A qualitative analysis of leadership styles and the selection of quality improvement in primary care practice

In the Practice Change Model, physicians act as key stakeholders, people who have both an investment in the practice and the capacity to influence how the practice performs. This leadership role is critical to the development and change of the practice. Leadership roles and effectiveness are an important factor in quality improvement in primary care practices.^ The study conducted involved a comparative case study analysis to identify leadership roles and the relationship between leadership roles and the number and type of quality improvement strategies adopted during a Practice Change Model-based intervention study. The research utilized secondary data from four primary care practices with various leadership styles. The practices are located in the San Antonio region and serve a large Hispanic population. The data was collected by two ABC Project Facilitators from each practice during a 12-month period including Key Informant Interviews (all staff members), MAP (Multi-method Assessment Process), and Practice Facilitation field notes. This data was used to evaluate leadership styles, management within the practice, and intervention tools that were implemented. The chief steps will be (1) to analyze if the leader-member relations contribute to the type of quality improvement strategy or strategies selected (2) to investigate if leader-position power contributes to the number of strategies selected and the type of strategy selected (3) and to explore whether the task structure varies across the four primary care practices.^ The research found that involving more members of the clinic staff in decision-making, building bridges between organizational staff and clinical staff, and task structure are all associated with the direct influence on the number and type of quality improvement strategies implemented in primary care practice.^ Although this research only investigated leadership styles of four different practices, it will offer future guidance on how to establish the priorities and implementation of quality improvement strategies that will have the greatest impact on patient care improvement. ^

Functional data analysis approaches for genotype-phenotype association studies from next-generation sequencing

Next-generation DNA sequencing platforms can effectively detect the entire spectrum of genomic variation and is emerging to be a major tool for systematic exploration of the universe of variants and interactions in the entire genome. However, the data produced by next-generation sequencing technologies will suffer from three basic problems: sequence errors, assembly errors, and missing data. Current statistical methods for genetic analysis are well suited for detecting the association of common variants, but are less suitable to rare variants. This raises great challenge for sequence-based genetic studies of complex diseases.^ This research dissertation utilized genome continuum model as a general principle, and stochastic calculus and functional data analysis as tools for developing novel and powerful statistical methods for next generation of association studies of both qualitative and quantitative traits in the context of sequencing data, which finally lead to shifting the paradigm of association analysis from the current locus-by-locus analysis to collectively analyzing genome regions.^ In this project, the functional principal component (FPC) methods coupled with high-dimensional data reduction techniques will be used to develop novel and powerful methods for testing the associations of the entire spectrum of genetic variation within a segment of genome or a gene regardless of whether the variants are common or rare.^ The classical quantitative genetics suffer from high type I error rates and low power for rare variants. To overcome these limitations for resequencing data, this project used functional linear models with scalar response to develop statistics for identifying quantitative trait loci (QTLs) for both common and rare variants. To illustrate their applications, the functional linear models were applied to five quantitative traits in Framingham heart studies. ^ This project proposed a novel concept of gene-gene co-association in which a gene or a genomic region is taken as a unit of association analysis and used stochastic calculus to develop a unified framework for testing the association of multiple genes or genomic regions for both common and rare alleles. The proposed methods were applied to gene-gene co-association analysis of psoriasis in two independent GWAS datasets which led to discovery of networks significantly associated with psoriasis.^

Are anatomic brain locations important in determining the clinical outcomes in acute ischemic stroke? --- An exploratory analysis of the NINDS Stroke Trial data

Background and purpose. Brain lesions in acute ischemic stroke measured by imaging tools provide important clinical information for diagnosis and final infarct volume has been considered as a potential surrogate marker for clinical outcomes. Strong correlations have been found between lesion volume and clinical outcomes in the NINDS t-PA Stroke Trial but little has been published about lesion location and clinical outcomes. Studies of the National Institute of Neurological Disorders and Stroke (NINDS) t-PA Stroke Trial data found the direction of the t-PA treatment effect on a decrease in CT lesion volume was consistent with the observed clinical effects at 3 months, but measure of t-PA treatment benefits using CT lesion volumes showed a diminished statistical significance, as compared to using clinical scales. ^ Methods. We used the global test to evaluate the hypothesis that lesion locations were strongly associated with clinical outcomes within each treatment group at 3 months after stroke. The anatomic locations of CT scans were used for analysis. We also assessed the effect of t-PA on lesion location using a global statistical test. ^ Results. In the t-PA group, patients with frontal lesions had larger infarct volumes and worse NIHSS score at 3 months after stroke. The clinical status of patients with frontal lesions in t-PA group was less likely to be affected by lesion volume, as compared to those who had no frontal lesions in at 3 months. For patients within the placebo group, both brain stem and internal capsule locations were significantly associated with a lower odd of having favorable outcomes at 3 months. Using a global test we could not detect a significant effect of t-PA treatment on lesion location although differences between two treatment groups in the proportion of lesion findings in each location were found. ^ Conclusions. Frontal, brain stem, and internal capsule locations were significantly related to clinical status at 3 months after stroke onset. We detect no significant t-PA effect on all 9 locations although proportion of lesion findings in differed among locations between the two treatment groups.^

TIME SERIES ANALYSIS AS INPUT FOR PREDICTIVE MODELING: PREDICTING CARDIAC ARREST IN A PEDIATRIC INTENSIVE CARE UNIT

The first manuscript, entitled "Time-Series Analysis as Input for Clinical Predictive Modeling: Modeling Cardiac Arrest in a Pediatric ICU" lays out the theoretical background for the project. There are several core concepts presented in this paper. First, traditional multivariate models (where each variable is represented by only one value) provide single point-in-time snapshots of patient status: they are incapable of characterizing deterioration. Since deterioration is consistently identified as a precursor to cardiac arrests, we maintain that the traditional multivariate paradigm is insufficient for predicting arrests. We identify time series analysis as a method capable of characterizing deterioration in an objective, mathematical fashion, and describe how to build a general foundation for predictive modeling using time series analysis results as latent variables. Building a solid foundation for any given modeling task involves addressing a number of issues during the design phase. These include selecting the proper candidate features on which to base the model, and selecting the most appropriate tool to measure them. We also identified several unique design issues that are introduced when time series data elements are added to the set of candidate features. One such issue is in defining the duration and resolution of time series elements required to sufficiently characterize the time series phenomena being considered as candidate features for the predictive model. Once the duration and resolution are established, there must also be explicit mathematical or statistical operations that produce the time series analysis result to be used as a latent candidate feature. In synthesizing the comprehensive framework for building a predictive model based on time series data elements, we identified at least four classes of data that can be used in the model design. The first two classes are shared with traditional multivariate models: multivariate data and clinical latent features. Multivariate data is represented by the standard one value per variable paradigm and is widely employed in a host of clinical models and tools. These are often represented by a number present in a given cell of a table. Clinical latent features derived, rather than directly measured, data elements that more accurately represent a particular clinical phenomenon than any of the directly measured data elements in isolation. The second two classes are unique to the time series data elements. The first of these is the raw data elements. These are represented by multiple values per variable, and constitute the measured observations that are typically available to end users when they review time series data. These are often represented as dots on a graph. The final class of data results from performing time series analysis. This class of data represents the fundamental concept on which our hypothesis is based. The specific statistical or mathematical operations are up to the modeler to determine, but we generally recommend that a variety of analyses be performed in order to maximize the likelihood that a representation of the time series data elements is produced that is able to distinguish between two or more classes of outcomes. The second manuscript, entitled "Building Clinical Prediction Models Using Time Series Data: Modeling Cardiac Arrest in a Pediatric ICU" provides a detailed description, start to finish, of the methods required to prepare the data, build, and validate a predictive model that uses the time series data elements determined in the first paper. One of the fundamental tenets of the second paper is that manual implementations of time series based models are unfeasible due to the relatively large number of data elements and the complexity of preprocessing that must occur before data can be presented to the model. Each of the seventeen steps is analyzed from the perspective of how it may be automated, when necessary. We identify the general objectives and available strategies of each of the steps, and we present our rationale for choosing a specific strategy for each step in the case of predicting cardiac arrest in a pediatric intensive care unit. Another issue brought to light by the second paper is that the individual steps required to use time series data for predictive modeling are more numerous and more complex than those used for modeling with traditional multivariate data. Even after complexities attributable to the design phase (addressed in our first paper) have been accounted for, the management and manipulation of the time series elements (the preprocessing steps in particular) are issues that are not present in a traditional multivariate modeling paradigm. In our methods, we present the issues that arise from the time series data elements: defining a reference time; imputing and reducing time series data in order to conform to a predefined structure that was specified during the design phase; and normalizing variable families rather than individual variable instances. The final manuscript, entitled: "Using Time-Series Analysis to Predict Cardiac Arrest in a Pediatric Intensive Care Unit" presents the results that were obtained by applying the theoretical construct and its associated methods (detailed in the first two papers) to the case of cardiac arrest prediction in a pediatric intensive care unit. Our results showed that utilizing the trend analysis from the time series data elements reduced the number of classification errors by 73%. The area under the Receiver Operating Characteristic curve increased from a baseline of 87% to 98% by including the trend analysis. In addition to the performance measures, we were also able to demonstrate that adding raw time series data elements without their associated trend analyses improved classification accuracy as compared to the baseline multivariate model, but diminished classification accuracy as compared to when just the trend analysis features were added (ie, without adding the raw time series data elements). We believe this phenomenon was largely attributable to overfitting, which is known to increase as the ratio of candidate features to class examples rises. Furthermore, although we employed several feature reduction strategies to counteract the overfitting problem, they failed to improve the performance beyond that which was achieved by exclusion of the raw time series elements. Finally, our data demonstrated that pulse oximetry and systolic blood pressure readings tend to start diminishing about 10-20 minutes before an arrest, whereas heart rates tend to diminish rapidly less than 5 minutes before an arrest.

Renal Cryoablation: Investigation of Periprocedural Visualization Tools and Treatment Response Quantification

Cryoablation for small renal tumors has demonstrated sufficient clinical efficacy over the past decade as a non-surgical nephron-sparing approach for treating renal masses for patients who are not surgical candidates. Minimally invasive percutaneous cryoablations have been performed with image guidance from CT, ultrasound, and MRI. During the MRI-guided cryoablation procedure, the interventional radiologist visually compares the iceball size on monitoring images with respect to the original tumor on separate planning images. The comparisons made during the monitoring step are time consuming, inefficient and sometimes lack the precision needed for decision making, requiring the radiologist to make further changes later in the procedure. This study sought to mitigate uncertainty in these visual comparisons by quantifying tissue response to cryoablation and providing visualization of the response during the procedure. Based on retrospective analysis of MR-guided cryoablation patient data, registration and segmentation algorithms were investigated and implemented for periprocedural visualization to deliver iceball position/size with respect to planning images registered within 3.3mm with at least 70% overlap and a quantitative logit model was developed to relate perfusion deficit in renal parenchyma visualized in verification images as a result of iceball size visualized in monitoring images. Through retrospective study of 20 patient cases, the relationship between likelihood of perfusion loss in renal parenchyma and distance within iceball was quantified and iteratively fit to a logit curve. Using the parameters from the logit fit, the margin for 95% perfusion loss likelihood was found to be 4.28 mm within the iceball. The observed margin corresponds well with the clinically accepted margin of 3-5mm within the iceball. In order to display the iceball position and perfusion loss likelihood to the radiologist, algorithms were implemented to create a fast segmentation and registration module which executed in under 2 minutes, within the clinically-relevant 3 minute monitoring period. Using 16 patient cases, the average Hausdorff distance was reduced from 10.1mm to 3.21 mm with average DSC increased from 46.6% to 82.6% before and after registration.

A comparative analysis of hospital -based physician monitoring systems

The purpose of the multiple case-study was to determine how hospital subsystems (such as physician monitoring and credentialing; quality assurance; risk management; and peer review) were supporting the monitoring of physicians? Three large metropolitan hospitals in Texas were studied and designated as hospitals #1, #2, and #3. Realizing that hospital subsystems are a unique entity and part of a larger system, conclusions were made on the premises of a quality control system, in relation to the tools of government (particularly the Health Care Quality Improvement Act (HCQIA)), and in relation to itself as a tool of a hospital.^ Three major analytical assessments were performed. First, the subsystems were analyzed as to their "completeness"; secondly, the subsystems were analyzed for "performance"; and thirdly, the subsystems were analyzed in reference to the interaction of completeness and performance.^ The physician credentialing and monitoring and the peer review subsystems as quality control systems were most complete, efficient, and effective in hospitals #1 and #3. The HCQIA did not seem to be an influencing factor in the completeness of the subsystem in hospital #1. The quality assurance and risk management subsystem in hospital #2 was not representative of completeness and performance and the HCQIA was not an influencing factor in the completeness of the Q.A. or R.M. systems in any hospital. The efficiency (computerization) of the physician credentialing, quality assurance and peer review subsystems in hospitals #1 and #3 seemed to contribute to their effectiveness (system-wide effect).^ The results indicated that the more complete, effective, and efficient subsystems were characterized by (1) all defined activities being met, (2) the HCQIA being an influencing factor, (3) a decentralized administrative structure, (4) computerization an important element, and (5) staff was sophisticated in subsystem operations. However, other variables were identified which deserve further research as to their effect on completeness and performance of subsystems. They include (1) medical staff affiliations, (2) system funding levels, (3) the system's administrative structure, and (4) the physician staff "cultural" characteristics. Perhaps by understanding other influencing factors, health care administrators may plan subsystems that will be compatible with legislative requirements and administrative objectives. ^