Search

There is no single test clinically available that is independently diagnostic for multiple sclerosis (MS). Currently MS is diagnosed using a combination of clinical evaluation and investigations including magnetic resonance imaging (MRI), interpreted in accordance with diagnostic criteria, to demonstrate the requisite dissemination of lesions in (anatomical) space and time. Lesions comprising inflammatory demyelination in the central nervous system are a core pathological feature of MS. Ultra-high field (e.g. 7 Tesla or 7T) T2*-weighted MRI can demonstrate in vivo a central vein in most of these lesions. This is a histopathologically specific feature which could be exploited to improve diagnostic workup in cases of suspected inflammatory demyelination. Central nervous system white matter not involved in demyelinating lesions is nevertheless affected in MS. The mechanisms inflicting injury to this normal appearing white matter (NAWM) and how they relate to focal lesions are unclear. Damage to NAWM seems important, because it correlates well with disability. Any association between cortical lesions, focal white matter lesions (WML) and diffuse damage to NAWM is difficult to investigate in vivo in MS, principally because MRI is relatively insensitive to cortical lesions. Investigation of such associations may also be confounded by the presence of small focal lesions within the “NAWM” that may remain undetected when using conventional MRI to define NAWM. Advantages inherent to ultra-high field MRI might help mitigate both of these problems.

This thesis addresses the three principal questions concerning the development of CT urography for investigating haematuria and each question is the subject of a separate chapter. The questions are: What is the reasoning behind using CT urography? What is the optimum diagnostic strategy using CT urography? What are the problems with using CT urography and how may solutions be provided? Haematuria can signify serious disease such as urinary tract stones, renal cell cancer, upper tract urothelial cancer (UTUC) and bladder cancer (BCa). CT urography is defined as contrast enhanced CT examination of kidneys, ureters and bladder. The technique used here includes unenhanced, nephrographic and excretory-phases for optimized diagnosis of stones, renal masses and urothelial cancer respectively. The reasoning behind using excretory-phase CT urography for investigating haematuria is based on results showing its high diagnostic accuracy for UTUC and BCa. Patients with haematuria are classified as low risk or high risk for UTUC and BCa, by a risk score, determined by the presence/absence of risk factors: age > 50 years, visible or nonvisible haematuria, history of smoking and occupational exposure. The optimum diagnostic strategy for patients at high risk for urothelial cancer, uses CT urography as a replacement test for ultrasonography and intravenous urography and as a triage test for flexible and rigid cystoscopy, resulting in earlier diagnosis and potentially improving prognosis. For patients at low risk, ultrasonography, unenhanced and nephrographic-phase CT urography are proposed as initial imaging tests. Problems with using CT urography include false positive results for UTUC, which are eliminated by retrograde ureteropyelography-guided biopsy, an innovative technique, for histopathological confirmation of diagnosis. Recommendations for the NHS and possible future developments are discussed. CT urography, including excretory-phase imaging, is recommended as the initial diagnostic imaging test before cystoscopy for patients with haematuria at high risk for urothelial cancer.

Background: Diagnosis of acute coronary syndromes (ACS) in the Emergency Department (ED) is a topical and contentious issue. Current diagnostic techniques rely on hospital admission for troponin testing. Only a minority of those admitted prove to have ACS while unacceptable proportions of those discharged have unrecognised ACS. Aims: We aimed to evaluate the diagnostic and prognostic value of individual clinical findings and novel biomarkers in ED patients with suspected cardiac chest pain. We then aimed to derive a clinical decision rule (CDR) to potentially enable safe, immediate discharge of a proportion of patients from the ED while risk stratifying others to facilitate triage to an appropriate level of in-patient care. Methods: We recruited patients who presented to the ED with suspected cardiac chest pain. Variables that have previously been shown to predict diagnosis of acute myocardial infarction (AMI) or to predict outcome were prospectively recorded. Blood was drawn at presentation for levels of eight biomarkers. Patients underwent 12-hour troponin testing and were followed up for the composite primary outcome of AMI, death or urgent coronary revascularisation for six months. Variables that were univariate predictors (p<0.05) of outcome were entered into a multivariate analysis using recursive partitioning. Results: While many clinical findings and levels of all eight novel biomarkers were found to be significant predictors of outcome, none could be used individually to confirm or exclude ACS in the ED. We derived a nine-point CDR that combined clinical findings with biomarker levels to effectively stratify patients into four risk groups. 14.2% of patients were identified as being at ‘no risk’ and had a 0.0% outcome rate. The rule performed significantly better than two commonly used risk scores and may improve on triage decisions made in actual clinical practice. Conclusion: ACS remains a difficult diagnosis to confidently confirm or refute in the ED. Our CDR may help to avoid unnecessary hospital admissions while improving on triage decisions made for the remaining in-patients. Prospective validation of our findings is warranted.

Objective: To evaluate the accuracy of a questionnaire designed for the identification of intraoral pain with neuropathic characteristics in a clinical orofacial pain sample population. Methods: 136 participants with 4 orofacial pain diagnoses (temporomandibular disorders [TMD, n=41], acute dental pain [ADP, n=41], trigeminal neuralgia (TN, n=19), persistent dentoalveolar pain disorder [PDAP, n=14]) and a group of pain-free controls (n=21) completed the modified S-LANSS, an adaptation of the original questionnaire devised to detected patients suffering from intraoral pain with neuropathic characteristics. Psychometric properties (sensitivity, specificity, positive predictive value [PPV], negative predictive value [NPV]) were calculated in two analyses with two different thresholds: 1) Detection of pain with neuropathic characteristics: PDAP + TN were considered positive, and TMD + ADP + controls were considered negative per gold standard (expert opinion). 2) Detection of PDAP: PDAP was considered positive and TMD + ADP were considered negative per gold standard. For both analyses, target values for adequate accuracy were defined as ≥80% for sensitivity and specificity. Results: For detection of orofacial pain with neuropathic characteristics (PDAP + TN), the modified S-LANSS presented with the most optimistic threshold sensitivity of 52% (95% CI: 34-69%), specificity of 70% (95% CI: 60-79%), PPV of 35% (95% CI: 22-51%), and NPV of 82% (95% CI: 72-89%). For detection of PDAP only, with the most optimistic threshold sensitivity was 64% (95% CI: 35-87%), specificity 63% (95% CI: 52-74%), PPV 23% (95% CI: 11-39%) and NPV 91% (95% CI: 81-97%). Conclusion: Based on a priori defined criteria, the modified S-LANSS did not show adequate accuracy to detect intraoral pain with neuropathic characteristics in a clinical orofacial pain population.

Objectives: Two common types of orofacial pains that are clinically important to distinguish are odontogenic pain and temporomandibular disorders (TMD) pain. The aim of this study was to determine the sensitivity and specificity of two screening instruments in distinguishing patients with odontogenic pain from patients with TMD pain. Methods: A convenience sample of patients seeking care at an Endodontist’s office and at an Orofacial Pain Clinic was recruited. The 14-item Dental Pain Questionnaire (DePaQ) (Pau et al., 2005) was used to screen for odontogenic pain and the 6-item TMD-screener (Gonzalez et al., 2011) was used to screen for TMD pain. Sensitivity and specificity calculations, with 95% confidence intervals, were performed for both instruments. Results: Thirty-four patients with odontogenic pain and 37 patients with TMD pain were enrolled. Both groups had comparable age distribution (49±12 vs. 45±18 years) and gender composition (53% vs. 86% females). Results of the sensitivity and specificity are provided in the table below with determination of diagnostic accuracy of these measures evaluated using published guidelines. Odontogenic Pain versus TMD Pain (95% confidence interval) DePaQ Sensitivity 85% (69% to 95%) Specificity 11% (3% to 25%) TMD-screener Sensitivity 92% (78% to 98%) Specificity 59% (41% to 75%) Conclusion: The point estimates for both the DePaQ and TMD screener were “acceptable” in identifying patients who had the pain condition in question (i.e., sensitivity), although the point estimate for appropriately identifying patients who did not have the pain condition when they did not have it (i.e., specificity) was “non-acceptable” neither for the DePaQ nor for the TMD screener. The potential for high amounts of false positive responses with the DePaQ limits its use as a screening instrument when patients are suspected as having TMD pain.

This dissertation consists of three projects in the area of group testing. The method of group testing, through the use of pooling, has proven to be an efficient method of reducing the time and cost associated with screening for a binary characteristic of interest, such as infection status. The salient feature of group testing that provides for these gains in efficiency is that testing is performed on pooled specimens, rather than testing specimens one-by-one. In Chapter 1, we present a general introduction of group testing. Typically, the statistical literature surrounding group testing has investigated the implementation of pooled testing for the purposes of either case identification or estimation. In this dissertation, we mainly focuses on the estimation problem which involves the development of regression models that relate individual level covariates to testing responses observed from pooled specimens. Primarily, the existing research in the area of estimation in group testing has focused on parametric regression models, where the shape of the link function is assumed as known and only a finite number of regression parameters has to be estimated. Recently, for the purpose of obviating the specification of the link function and increasing the flexibility of modeling, nonparametric group testing regression models have been studied. %It considers the case where each individual has one continuous explanatory variable and the link function is a univariate probability curve. Existing methods of estimating this unknown function are based on local moment estimators. In Chapter 2, we propose a new nonparametric estimation procedure using a local likelihood approach. For easy illustration, in this part we consider the situation where each individual is assigned to exactly one pool and only this pooled specimen is tested. Further, we assume the assay used for screening is perfect. Both of these two assumptions will be relaxed in the rest chapters of this dissertation. We show that our proposed estimator enjoys an asymptotic normal distribution with the optimal nonparametric estimation rate. Finite sample performance of the method is exhibited via some simulated examples and a real data analysis. To pursue a more suitable technique of modeling group testing data, in Chapter 3, we develop a general semiparametric framework which allows for the inclusion of only not one continuous covariate, but also multiple explanatory variables, all variants of decoding information, and imperfect testing. The asymptotic properties of our estimators are presented and guidance on finite sample implementation is provided. We illustrate the performance of our methods through simulation and by applying them to chlamydia and gonorrhea data collected by the Nebraska Public Health Laboratory as a part of the Infertility Prevention Project. In Chapter 4, we focus on the evaluation of misclassification effect of testing pools which are constructed according to any types of group testing algorithms. The existing assumption regarding them are somehow restrictive. If they are invalid, the estimation procedure can lead to severely biased estimator. In this work, we relax previously made assumptions regarding testing error rates by acknowledging the underlying mechanistic structure of the diagnostic test being employed. For easy illustration of this methodology, we mainly concentrate in parametric regression methods and propose a general estimation framework that allows for the analysis of data arising from all group testing strategies. The finite sample performance of our proposed methodology are investigated through simulation and by applying our techniques to hepatitis B data from a study involving Irish prisoners. Through these studies, we show that our methods can result in more efficient parameter estimates, when compared to competing procedures that make use of individual level data, at a fraction of the cost of data collection. Before proceeding to the main body of this dissertation, I would like to clarify that the notations defined in this work are self-contained in each separated chapter.

Metaplastic breast carcinoma (MBC) is a rare subtype of breast cancer characterized by coexistence of carcinomatous and sarcomatous components. Snail is a nuclear transcription factor incriminated in the transition of epithelial to mesenchymal differentiation of breast cancer. Aberrant Snail expression results in lost expression of the cell adhesion molecule E-cadherin, an event associated with changes in epithelial architecture and invasive growth. We aimed to identify the utility of Snail, and of traditional immunohistochemical markers, in accurate MBC classification and to evaluate clinicopathologic characteristics and outcome.We retrospectively reviewed 34 MBC cases from January 1997 to September 2007. The control group contained 26 spindle cell lesions. Immunohistochemistry used Snail, p63, epidermal growth factor receptor (EGFR), OSCAR, and wide spectrum cytokeratin (WS-KER). Negative was a score less than 1%. We found that Snail and EGFR are sensitive (100%) markers with low specificity (3.8% and 19.2%) for detecting MBC. p63 and WS-KER are specific (100%), with moderate sensitivity (67.6% and 76.5%); OSCAR is sensitive (85.3%) and specific (92.3%). A combination of any 2 of the p63, OSCAR, and WS-KER markers increased sensitivity and specificity. MBCs tended to be high-grade (77%), triple negative (negative for estrogen receptor, progesterone receptor, and HER2) [27/33; 81.8%], and carcinomas with low incidence of axillary lymph node involvement (15%), and decreased disease-free [71% (95%CI: 54%, 94%) at 3 yrs.) and overall survival. A combination of p63, OSCAR and WS-KER are useful in its work-up. On the other hand, Snail is neither a diagnostic nor a prognostic marker for MBC.

Fourier Transform Infrared (FTIR) microscopy is a sensitive method for detecting difference in the morphology of biological cells. In this study FTIR spectra were obtained for uninfected cells, and cells infected with two different viruses. The spectra obtained are difficult to discriminate visually. Here we apply advanced statistical methods to the analysis of the spectra, to test if such spectra are useful for diagnosing viral infections in cells. Logistic Regression (LR) and Partial Least Squares Regression (PLSR) were used to build models which allow us to diagnose if spectral differences are related to infection state of the cells. A three-fold, balanced cross-validation method was applied to estimate the shrinkages of the area under the receiving operator characteristic curve (AUC), and specificities at sensitivities of 95%, 90% and 80%. AUC, sensitivity and specificity were used to gauge the goodness of the discrimination methods. Our statistical results shows that the spectra associated with different cellular states are very effectively discriminated. We also find that the overall performance of PLSR is better than that of LR, especially for new data validation. Our analysis supports the idea that FTIR microscopy is a useful tool for detection of viral infections in biological cells.

The aim of this study was to establish which of several commonly used parameters performs best in the evaluation of adnexal masses by transvaginal color Doppler sonography. A total of 79 adnexal masses in 73 consecutive patients were included in the study. There were 43 (58.9%) premenopausal and 30 (41.1%) postmenopausal patients. The median age was 45 years (range 20-78 years). The parameters compared were: number of vessels detected in each tumor, tumor vessel location (central vs. peripheral), peak systolic velocity (PSV), lowest resistance index (RIlowest), mean resistance index (RImean), lowest pulsatility index (PIlowest) and mean pulsatility index (PImean). Receiver operating characteristic (ROC) curves were plotted to analyze the test performance of the parameters, except for tumor vessel location, and to estimate the best cut-off value of the parameters studied to differentiate between malignant and benign tumors. Definitive histopathological diagnosis was obtained in every case and used as the 'gold standard'. There were 20 (25.3%) malignant and 59 (74.7%) benign masses. Color Doppler signals were detected in 100% of the malignant masses (20 out of 20) and 74.6% (44 out of 59) of the benign masses, and the difference was found to be statistically significant (p < 0.001). Tumor vessel location was central in 18 out of 20 (90%) malignant masses, whereas it was peripheral in 39 out of 44 (88.6%) benign masses. ROC analysis showed that the best cut-off values for number of vessels, PSV, RIlowest, RImean, PIlowest and PImean were three vessels, 25 cm/s, 0.45, 0.55, 0.90 and 1.50, respectively. However, for all these parameters except RIlowest, there was a considerable overlap between benign and malignant tumors, with a high false-positive rate. In conclusion, in our experience, the parameters that performed best were the RIlowest with a cut-off value of 0.45 (sensitivity 100%; false-positive rate 11.4%) and central tumor vessel location (sensitivity 90%; false-positive rate 11.4%).