Towards greater accuracy in individual-tree mortality regression

Background mortality is an essential component of any forest growth and yield model. Forecasts of mortality contribute largely to the variability and accuracy of model predictions at the tree, stand and forest level. In the present study, I implement and evaluate state-of-the-art techniques to increase the accuracy of individual tree mortality models, similar to those used in many of the current variants of the Forest Vegetation Simulator, using data from North Idaho and Montana. The first technique addresses methods to correct for bias induced by measurement error typically present in competition variables. The second implements survival regression and evaluates its performance against the traditional logistic regression approach. I selected the regression calibration (RC) algorithm as a good candidate for addressing the measurement error problem. Two logistic regression models for each species were fitted, one ignoring the measurement error, which is the “naïve” approach, and the other applying RC. The models fitted with RC outperformed the naïve models in terms of discrimination when the competition variable was found to be statistically significant. The effect of RC was more obvious where measurement error variance was large and for more shade-intolerant species. The process of model fitting and variable selection revealed that past emphasis on DBH as a predictor variable for mortality, while producing models with strong metrics of fit, may make models less generalizable. The evaluation of the error variance estimator developed by Stage and Wykoff (1998), and core to the implementation of RC, in different spatial patterns and diameter distributions, revealed that the Stage and Wykoff estimate notably overestimated the true variance in all simulated stands, but those that are clustered. Results show a systematic bias even when all the assumptions made by the authors are guaranteed. I argue that this is the result of the Poisson-based estimate ignoring the overlapping area of potential plots around a tree. Effects, especially in the application phase, of the variance estimate justify suggested future efforts of improving the accuracy of the variance estimate. The second technique implemented and evaluated is a survival regression model that accounts for the time dependent nature of variables, such as diameter and competition variables, and the interval-censored nature of data collected from remeasured plots. The performance of the model is compared with the traditional logistic regression model as a tool to predict individual tree mortality. Validation of both approaches shows that the survival regression approach discriminates better between dead and alive trees for all species. In conclusion, I showed that the proposed techniques do increase the accuracy of individual tree mortality models, and are a promising first step towards the next generation of background mortality models. I have also identified the next steps to undertake in order to advance mortality models further.

Determination of carbohydrate-deficient transferrin in human serum by two capillary zone electrophoresis methods and a direct immunoassay: comparison of patient data

Data obtained with two CZE assays for determining carbohydrate-deficient transferrin (CDT) in human serum under routine conditions, the CAPILLARYS CDT and the high-resolution CEofix (HR-CEofix) CDT methods, are in agreement with patient sera that do not exhibit interferences, high trisialo-transferrin (Tf) levels or genetic variants. HR-CEofix CDT levels are somewhat higher compared to those obtained with the CAPILLARYS method and this bias corresponds to the difference of the upper reference values of the two assays. The lower resolution between disialo-Tf and trisialo-Tf observed in the CAPILLARYS system (mean: 1.24) compared to HR-CEofix (mean: 1.74) is believed to be the key for this difference. For critical sera with high trisialo-Tf levels, genetic variants, or certain interferences in the beta-region, the HR-CEofix approach is demonstrated to perform better than CAPILLARYS. However, the determination of CDT with the HR-CEofix method can also be hampered with interferences. Results with disialo-Tf values larger than 3% in the absence of asialo-Tf should be evaluated with immunosubtraction of Tf and possibly also confirmed with another CZE method or by HPLC. Furthermore, data gathered with the N Latex CDT direct immunonephelometric assay suggest that this assay can be used for screening purposes. To reduce the number of false negative results, CDT data above 2.0% should be confirmed using a separation method.

Validation of a new method for determination of cup orientation in THA

Our goal was to validate accuracy, consistency, and reproducibility/reliability of a new method for determining cup orientation in total hip arthroplasty (THA). This method allows matching the 3D-model from CT images or slices with the projected pelvis on an anteroposterior pelvic radiograph using a fully automated registration procedure. Cup orientation (inclination and anteversion) is calculated relative to the anterior pelvic plane, corrected for individual malposition of the pelvis during radiograph acquisition. Measurements on blinded and randomized radiographs of 80 cadaver and 327 patient hips were investigated. The method showed a mean accuracy of 0.7 +/- 1.7 degrees (-3.7 degrees to 4.0 degrees) for inclination and 1.2 +/- 2.4 degrees (-5.3 degrees to 5.6 degrees) for anteversion in the cadaver trials and 1.7 +/- 1.7 degrees (-4.6 degrees to 5.5 degrees) for inclination and 0.9 +/- 2.8 degrees (-5.2 degrees to 5.7 degrees) for anteversion in the clinical data when compared to CT-based measurements. No systematic errors in accuracy were detected with the Bland-Altman analysis. The software consistency and the reproducibility/reliability were very good. This software is an accurate, consistent, reliable, and reproducible method to measure cup orientation in THA using a sophisticated 2D/3D-matching technique. Its robust and accurate matching algorithm can be expanded to statistical models.

Determination of ethyl glucuronide in human serum by capillary zone electrophoresis and an immunoassay

Ethyl glucuronide (EtG) is a marker of recent alcohol consumption. For the optimization of the analysis of EtG by CZE with indirect absorbance detection, the use of capillaries with permanent and dynamic wall coatings, the composition of the BGE, and various sample preparation procedures, including dilution with water, ultrafiltration, protein precipitation, and SPE, were investigated. Two validated screening assays for the determination of EtG in human serum, a CZE-based approach and an enzyme immunoassay (EIA), are described. The CZE assay uses a coated capillary, 2,4-dimethylglutaric acid as an internal standard, and a pH 4.65 BGE comprising 9 mM nicotinic acid, epsilon-aminocaproic acid and 10% v/v ACN. Proteins are removed via precipitation with ACN prior to analysis and the LOQ is 0.50 mg/L. The EIA is based upon commercial reagents which are promoted for the determination of urinary EtG. Krebs-Ringer solution containing 5% BSA is used as a calibration matrix. All samples are ultrafiltered prior to analysis of the ultrafiltrate on a Mira Plus analyzer. Assay calibration ranged between 0 and 2 mg/L and the upper reference limit was determined to be 0.05 mg/L. Both assays proved to be suitable for the analysis of samples from different individuals. For EtG levels above 0.50 mg/L, good agreement was observed for the comparison of the results of the two methods.

Three-dimensional imaging core laboratory of the endovascular aneurysm repair trials: validation of methodology

OBJECTIVE: The aim of this study was to establish and validate a three-dimensional imaging protocol for the assessment of Computed Tomography (CT) scans of abdominal aortic aneurysms in UK EVAR trials patients. Quality control and repeatability of anatomical measurements is important for the validity of any core laboratory. METHODS: Three different observers performed anatomical measurements on 50 preoperative CT scans of aortic aneurysms using the Vitrea 2 three-dimensional post-imaging software in a core laboratory setting. We assessed the accuracy of intra and inter observer repeatability of measurements, the time required for collection of measurements, 3 different levels of automation and 3 different automated criteria for measurement of neck length. RESULTS: None of the automated neck length measurements demonstrated sufficient accuracy and it was necessary to perform checking of the important automated landmarks. Good intra and limited inter observer agreement were achieved with three-dimensional assessment. Complete assessment of the aneurysm and iliacs took an average (SD) of 17.2 (4.1) minutes. CONCLUSIONS: Aortic aneurysm anatomy can be assessed reliably and quickly using three-dimensional assessment but for scans of limited quality, manual checking of important landmarks remains necessary. Using a set protocol, agreement between observers is satisfactory but not as good as within observers.

Validation of the estimation of physiologic ability and surgical stress (E-PASS) score in liver surgery

BACKGROUND: The estimation of physiologic ability and surgical stress (E-PASS) has been used to produce a numerical estimate of expected mortality and morbidity after elective gastrointestinal surgery. The aim of this study was to validate E-PASS in a selected cohort of patients requiring liver resections (LR). METHODS: In this retrospective study, E-PASS predictor equations for morbidity and mortality were applied to the prospective data from 243 patients requiring LR. The observed rates were compared with predicted rates using Fisher's exact test. The discriminative capability of E-PASS was evaluated using receiver-operating characteristic (ROC) curve analysis. RESULTS: The observed and predicted overall mortality rates were both 3.3% and the morbidity rates were 31.3 and 26.9%, respectively. There was a significant difference in the comprehensive risk scores for deceased and surviving patients (p = 0.043). However, the scores for patients with or without complications were not significantly different (p = 0.120). Subsequent ROC curve analysis revealed a poor predictive accuracy for morbidity. CONCLUSIONS: The E-PASS score seems to effectively predict mortality in this specific group of patients but is a poor predictor of complications. A new modified logistic regression might be required for LR in order to better predict the postoperative outcome.

Validation of a Western Blot for the detection of anti-Trichinella spp. antibodies in domestic pigs

Trichinellosis is a zoonotic disease in humans caused by Trichinella spp. According to international regulations and guidelines, serological surveillance can be used to demonstrate the absence of Trichinella spp. in a defined domestic pig population. Most enzyme-linked immunosorbent assay (ELISA) tests presently available do not yield 100% specificity, and therefore, a complementary test is needed to confirm the diagnosis of any initial ELISA seropositivity. The goal of the present study was to evaluate the sensitivity and specificity of a Western Blot assay based on somatic Trichinella spiralis muscle stage (L1) antigen using Bayesian modeling techniques. A total of 295 meat juice and serum samples from pigs negative for Trichinella larvae by artificial digestion, including 74 potentially cross-reactive sera of pigs with other nematode infections, and 93 meat juice samples from pigs infected with Trichinella larvae were included in the study. The diagnostic sensitivity and specificity of the Western Blot were ranged from 95.8% to 96.0% and from 99.5% to 99.6%, respectively. A sensitivity analysis showed that the model outcomes were hardly influenced by changes in the prior distributions, providing a high confidence in the outcomes of the models. This validation study demonstrated that the Western Blot is a suitable method to confirm samples that reacted positively in an initial ELISA.

[Field validation of a real-time PCR test for the detection of Mycoplasma hyopneumoniae in nasal swabs of live pigs]

Enzootic pneumonia (EP) of pigs, caused by Mycoplasma hyopneumoniae has been a notifiable disease in Switzerland since May 2003. The diagnosis of EP has been based on multiple methods, including clinical, bacteriological and epidemiological findings as well as pathological examination of lungs (mosaic diagnosis). With the recent development of a real-time PCR (rtPCR) assay with 2 target sequences a new detection method for M. hyopneumoniae became available. This assay was tested for its applicability to nasal swab material from live animals. Pigs from 74 herds (average 10 pigs per herd) were tested. Using the mosaic diagnosis, 22 herds were classified as EP positive and 52 as EP negative. From the 730 collected swab samples we were able to demonstrate that the rtPCR test was 100% specific. In cases of cough the sensitivity on herd level of the rtPCR is 100%. On single animal level and in herds without cough the sensitivity was lower. In such cases, only a positive result would be proof for an infection with M. hyopneumoniae. Our study shows that the rtPCR on nasal swabs from live pigs allows a fast and accurate diagnosis in cases of suspected EP.

Validation of novel 3-dimensional electrocardiographic mapping of atrial tachycardias by invasive mapping and ablation: a multicenter study

OBJECTIVES This study prospectively evaluated the role of a novel 3-dimensional, noninvasive, beat-by-beat mapping system, Electrocardiographic Mapping (ECM), in facilitating the diagnosis of atrial tachycardias (AT). BACKGROUND Conventional 12-lead electrocardiogram, a widely used noninvasive tool in clinical arrhythmia practice, has diagnostic limitations. METHODS Various AT (de novo and post-atrial fibrillation ablation) were mapped using ECM followed by standard-of-care electrophysiological mapping and ablation in 52 patients. The ECM consisted of recording body surface electrograms from a 252-electrode-vest placed on the torso combined with computed tomography-scan-based biatrial anatomy (CardioInsight Inc., Cleveland, Ohio). We evaluated the feasibility of this system in defining the mechanism of AT-macro-re-entrant (perimitral, cavotricuspid isthmus-dependent, and roof-dependent circuits) versus centrifugal (focal-source) activation-and the location of arrhythmia in centrifugal AT. The accuracy of the noninvasive diagnosis and detection of ablation targets was evaluated vis-à-vis subsequent invasive mapping and successful ablation. RESULTS Comparison between ECM and electrophysiological diagnosis could be accomplished in 48 patients (48 AT) but was not possible in 4 patients where the AT mechanism changed to another AT (n = 1), atrial fibrillation (n = 1), or sinus rhythm (n = 2) during the electrophysiological procedure. ECM correctly diagnosed AT mechanisms in 44 of 48 (92%) AT: macro-re-entry in 23 of 27; and focal-onset with centrifugal activation in 21 of 21. The region of interest for focal AT perfectly matched in 21 of 21 (100%) AT. The 2:1 ventricular conduction and low-amplitude P waves challenged the diagnosis of 4 of 27 macro-re-entrant (perimitral) AT that can be overcome by injecting atrioventricular node blockers and signal averaging, respectively. CONCLUSIONS This prospective multicenter series shows a high success rate of ECM in accurately diagnosing the mechanism of AT and the location of focal arrhythmia. Intraprocedural use of the system and its application to atrial fibrillation mapping is under way.

An integrated system for 3D hip joint reconstruction from 2D X-rays: a preliminary validation study

The acquisition of conventional X-ray radiographs remains the standard imaging procedure for the diagnosis of hip-related problems. However, recent studies demonstrated the benefit of using three-dimensional (3D) surface models in the clinical routine. 3D surface models of the hip joint are useful for assessing the dynamic range of motion in order to identify possible pathologies such as femoroacetabular impingement. In this paper, we present an integrated system which consists of X-ray radiograph calibration and subsequent 2D/3D hip joint reconstruction for diagnosis and planning of hip-related problems. A mobile phantom with two different sizes of fiducials was developed for X-ray radiograph calibration, which can be robustly detected within the images. On the basis of the calibrated X-ray images, a 3D reconstruction method of the acetabulum was developed and applied together with existing techniques to reconstruct a 3D surface model of the hip joint. X-ray radiographs of dry cadaveric hip bones and one cadaveric specimen with soft tissue were used to prove the robustness of the developed fiducial detection algorithm. Computed tomography scans of the cadaveric bones were used to validate the accuracy of the integrated system. The fiducial detection sensitivity was in the same range for both sizes of fiducials. While the detection sensitivity was 97.96% for the large fiducials, it was 97.62% for the small fiducials. The acetabulum and the proximal femur were reconstructed with a mean surface distance error of 1.06 and 1.01 mm, respectively. The results for fiducial detection sensitivity and 3D surface reconstruction demonstrated the capability of the integrated system for 3D hip joint reconstruction from 2D calibrated X-ray radiographs.

Patient-specific finite-element simulation of the human cornea: A clinical validation study on cataract surgery

The planning of refractive surgical interventions is a challenging task. Numerical modeling has been proposed as a solution to support surgical intervention and predict the visual acuity, but validation on patient specific intervention is missing. The purpose of this study was to validate the numerical predictions of the post-operative corneal topography induced by the incisions required for cataract surgery. The corneal topography of 13 patients was assessed preoperatively and postoperatively (1-day and 30-day follow-up) with a Pentacam tomography device. The preoperatively acquired geometric corneal topography – anterior, posterior and pachymetry data – was used to build patient-specific finite element models. For each patient, the effects of the cataract incisions were simulated numerically and the resulting corneal surfaces were compared to the clinical postoperative measurements at one day and at 30-days follow up. Results showed that the model was able to reproduce experimental measurements with an error on the surgically induced sphere of 0.38D one day postoperatively and 0.19D 30 days postoperatively. The standard deviation of the surgically induced cylinder was 0.54D at the first postoperative day and 0.38D 30 days postoperatively. The prediction errors in surface elevation and curvature were below the topography measurement device accuracy of ±5μm and ±0.25D after the 30-day follow-up. The results showed that finite element simulations of corneal biomechanics are able to predict post cataract surgery within topography measurement device accuracy. We can conclude that the numerical simulation can become a valuable tool to plan corneal incisions in cataract surgery and other ophthalmosurgical procedures in order to optimize patients' refractive outcome and visual function.

Evaluation of the diagnostic value of serologic microagglutination testing and a polymerase chain reaction assay for diagnosis of acute leptospirosis in dogs in a referral center.

OBJECTIVE To determine the diagnostic value of a serologic microagglutination test (MAT) and a PCR assay on urine and blood for the diagnosis of leptospirosis in dogs with acute kidney injury (AKI). DESIGN Cross-sectional study. Animals-76 dogs with AKI in a referral hospital (2008 to 2009). PROCEDURES Dogs' leptospirosis status was defined with a paired serologic MAT against a panel of 11 Leptospira serovars as leptospirosis-associated (n = 30) or nonleptospirosis-associated AKI (12). In 34 dogs, convalescent serologic testing was not possible, and leptospirosis status was classified as undetermined. The diagnostic value of the MAT single acute or convalescent blood sample was determined in dogs in which leptospirosis status could be classified. The diagnostic value of a commercially available genus-specific PCR assay was evaluated by use of 36 blood samples and 20 urine samples. RESULTS Serologic acute testing of an acute blood sample had a specificity of 100% (95% CI, 76% to 100%), a sensitivity of 50% (33% to 67%), and an accuracy of 64% (49% to 77%). Serologic testing of a convalescent blood sample had a specificity of 92% (65% to 99%), a sensitivity of 100% (87% to 100%), and an accuracy of 98% (88% to 100%). Results of the Leptospira PCR assay were negative for all samples from dogs for which leptospirosis status could be classified. CONCLUSIONS AND CLINICAL RELEVANCE Serologic MAT results were highly accurate for diagnosis of leptospirosis in dogs, despite a low sensitivity for early diagnosis. In this referral setting of dogs pretreated with antimicrobials, testing of blood and urine samples with a commercially available genus-specific PCR assay did not improve early diagnosis.

Multicentre validation of the Geneva Risk Score for hospitalised medical patients at risk of venous thromboembolism. Explicit ASsessment of Thromboembolic RIsk and Prophylaxis for Medical PATients in SwitzErland (ESTIMATE).

There is a need to validate risk assessment tools for hospitalised medical patients at risk of venous thromboembolism (VTE). We investigated whether a predefined cut-off of the Geneva Risk Score, as compared to the Padua Prediction Score, accurately distinguishes low-risk from high-risk patients regardless of the use of thromboprophylaxis. In the multicentre, prospective Explicit ASsessment of Thromboembolic RIsk and Prophylaxis for Medical PATients in SwitzErland (ESTIMATE) cohort study, 1,478 hospitalised medical patients were enrolled of whom 637 (43%) did not receive thromboprophylaxis. The primary endpoint was symptomatic VTE or VTE-related death at 90 days. The study is registered at ClinicalTrials.gov, number NCT01277536. According to the Geneva Risk Score, the cumulative rate of the primary endpoint was 3.2% (95% confidence interval [CI] 2.2-4.6%) in 962 high-risk vs 0.6% (95% CI 0.2-1.9%) in 516 low-risk patients (p=0.002); among patients without prophylaxis, this rate was 3.5% vs 0.8% (p=0.029), respectively. In comparison, the Padua Prediction Score yielded a cumulative rate of the primary endpoint of 3.5% (95% CI 2.3-5.3%) in 714 high-risk vs 1.1% (95% CI 0.6-2.3%) in 764 low-risk patients (p=0.002); among patients without prophylaxis, this rate was 3.2% vs 1.5% (p=0.130), respectively. Negative likelihood ratio was 0.28 (95% CI 0.10-0.83) for the Geneva Risk Score and 0.51 (95% CI 0.28-0.93) for the Padua Prediction Score. In conclusion, among hospitalised medical patients, the Geneva Risk Score predicted VTE and VTE-related mortality and compared favourably with the Padua Prediction Score, particularly for its accuracy to identify low-risk patients who do not require thromboprophylaxis.

Simple Estimation of Incident HIV Infection Rates in Notification Cohorts Based on Window Periods of Algorithms for Evaluation of Line-Immunoassay Result Patterns

Background Tests for recent infections (TRIs) are important for HIV surveillance. We have shown that a patient's antibody pattern in a confirmatory line immunoassay (Inno-Lia) also yields information on time since infection. We have published algorithms which, with a certain sensitivity and specificity, distinguish between incident (< = 12 months) and older infection. In order to use these algorithms like other TRIs, i.e., based on their windows, we now determined their window periods. Methods We classified Inno-Lia results of 527 treatment-naïve patients with HIV-1 infection < = 12 months according to incidence by 25 algorithms. The time after which all infections were ruled older, i.e. the algorithm's window, was determined by linear regression of the proportion ruled incident in dependence of time since infection. Window-based incident infection rates (IIR) were determined utilizing the relationship ‘Prevalence = Incidence x Duration’ in four annual cohorts of HIV-1 notifications. Results were compared to performance-based IIR also derived from Inno-Lia results, but utilizing the relationship ‘incident = true incident + false incident’ and also to the IIR derived from the BED incidence assay. Results Window periods varied between 45.8 and 130.1 days and correlated well with the algorithms' diagnostic sensitivity (R2 = 0.962; P<0.0001). Among the 25 algorithms, the mean window-based IIR among the 748 notifications of 2005/06 was 0.457 compared to 0.453 obtained for performance-based IIR with a model not correcting for selection bias. Evaluation of BED results using a window of 153 days yielded an IIR of 0.669. Window-based IIR and performance-based IIR increased by 22.4% and respectively 30.6% in 2008, while 2009 and 2010 showed a return to baseline for both methods. Conclusions IIR estimations by window- and performance-based evaluations of Inno-Lia algorithm results were similar and can be used together to assess IIR changes between annual HIV notification cohorts.

Determination of Δ9-tetrahydrocannabinolic acid A (Δ9-THCA-A) in whole blood and plasma by LC–MS/MS and application in authentic samples from drivers suspected of driving under the influence of cannabis

Delta-9-tetrahydrocannabinolic acid A (THCA-A) is the biosynthetic precursor of delta-9-tetrahydrocannabinol (THC) in cannabis plants, and has no psychotropic effects. THCA-A can be detected in blood and urine, and several metabolites have been identified. THCA-A was also shown to be incorporated in hair by side stream smoke to a minor extent, but incorporation via blood stream or sweat seems unlikely. The detection of THCA-A in biological fluids may serve as a marker for differentiating between the intake of prescribed THC medication – containing only pure THC – and cannabis products containing THC besides THC-acid A and other cannabinoids. However, the knowledge about its usefulness in forensic cases is very limited. The aim of the present work was the development of a reliable method for THCA-A determination in human blood or plasma using LC–MS/MS and application to cases of driving under the influence of drugs. Fifty eight (58) authentic whole blood and the respective plasma samples were collected from drivers suspected of driving under the influence of cannabis from the region of Bern (Switzerland). Samples were first tested for THC, 11-OH-THC and THC-COOH, and then additionally for THCA-A. For this purpose, the existing LC–MS/MS method was modified and validated, and found to be selective and linear over a range of 1.0 to 200 ng/mL (the correlation coefficients were above 0.9980 in all validation runs). Limit of detection (LOD) and limit of quantification (LOQ) were 0.3 ng/mL and 1.0 ng/mL respectively. Intra- and inter-assay accuracy were equal or better than 90% and intra- and inter-assay precision were equal or better than 11.1%. The mean extraction efficiencies were satisfactory being equal or higher than 85.4%. THCA-A was stable in whole blood samples after 3 freeze/thaw cycles and storage at 4 °C for 7 days. Re-injection (autosampler) stability was also satisfactory. THC was present in all blood samples with levels ranging from 0.7 to 51 ng/mL. THCA-A concentrations ranged from 1.0 to 496 ng/mL in blood samples and from 1.4 to 824 ng/mL in plasma samples. The plasma:blood partition coefficient had a mean value of 1.7 (±0.21, SD). No correlation was found between the degree of intoxication or impairment stated in the police protocols or reports of medical examinations and the detected THCA-A-concentration in blood.