KL-6 in ANCA-Associated Vasculitis Patients with and without ILD: A Machine Learning Approach

Background: ANCA-associated vasculitis (AAV) are small vessel vasculitis distinguished between microscopic polyangiitis (MPA) and granulomatosis with polyangiitis (GPA). The former may have interstitial lung disease (ILD) associated with high morbidity and mortality. Here, Krebs von den Lungen-6 (KL-6), a marker of fibrotic ILD, was assessed for distinguishing AAV patients with ILD from those without ILD, and whether its changes over time are correlated with disease activity. Materials and Methods: Thirteen AAV patients (eight females, mean age 61 ± 14.8 years) were enrolled: six MPA and six GPA. Serum samples were assayed for KL-6 concentrations (Fujirebio Europe, Belgium). To investigate potential binary classifiers for diagnosis of AAV-ILD, we constructed a regression decision tree model. Results: Higher serum KL-6 were in AAV-ILD compared with those without ILD (972.8 ± 398.5 vs 305.4 ± 93.9, p = 0.0040). Area under the receiver operating characteristics curve showed 100% of the diagnostic performance of KL-6 for identifying the ILD involvement (accuracy 91.7%) and the best cutoff value of 368 U/mL (sensitivity 100% and specificity 87.5%). The decision tree model showed a 33% improvement in class purity using a cut-off value of 513 U/mL to distinguish AAV patients with and without ILD. Stratifying AAV patients as MPA and GPA with and without ILD considering T0 and T1 KL-6, the model obtained an improvement of 40% for classifying GPA non-ILD with a T0 serum KL-6 cut-off value of 513 U/mL and a T1 KL-6 cut-off of 301 U/mL. A direct correlation was found between serum T0 KL-6 and T0 BVAS (r = 0.578, p = 0.044). Conclusion: Our multicenter study demonstrated KL-6 as a reliable, non-invasive, and easy-to-perform marker of ILD in AAV patients and its helpfulness for disease activity assessment. Changes in serum concentrations of KL-6 over time could be useful for monitoring AAV patients. Further study of KL-6 as a marker of response to therapy during long-term follow-up would also be worthwhile.

Should rapid antigen tests be government funded in Australia? An economic evaluation

Objective: Easy and equitable access to testing is a cornerstone of the public health response to COVID-19. Currently in Australia, testing using Polymerase Chain Reaction (PCR) tests for COVID-19 is free-to-the-user, but the public purchase their own Rapid Antigen Tests (RATs). We conduct an economic analysis of government-funded RATs in Australia. Design: An interactive decision tree model was developed to compare one policy in which government-funded RATs are free-to-the-user, and one in which individuals purchase their own RATs. The decision tree represents RAT and PCR testing pathways for a cohort of individuals without COVID-19-like symptoms, to estimate the likelihood of COVID-19 positive individuals isolating prior to developing symptoms and the associated costs of testing, from a government perspective. Data sources: Test costs and detection rates were informed by published studies, other input parameter values are unobservable and uncertain, for which a range of scenario analyses are presented. Data synthesis: Assuming 10% prevalence of COVID-19 in a cohort of 10,000 individuals who would use government-funded RATs, the model estimates an additional 464 individuals would isolate early at a cost to the government of around $52,000. Scenario analyses indicate that the incremental cost per additional COVID-19 positive individual isolating with no symptoms remains at a few hundred dollars at 5% prevalence, rising to $2,052 at 1% prevalence. Conclusions: Based on the presented decision tree model, even only minor reductions in COVID-19 transmission rates due to early isolation would justify the additional costs associated with a policy of government-funded RATs.

The Method of Real Options and the Business Model «Lean Canvas» in the Practice of Performance Evaluation of it Projects

The relevance of the study is due to the active implementation of IT technologies in various aspects of companies, which gives special importance to the development of a methodology for assessing the effectiveness of projects in a highly uncertain environment. The paper presents the methodology and assesses the effectiveness of IT projects using binomial «decision tree» model and iterative risk assessment metamodel «Lean Canvas». The comparative assessment of IT project efficiency using discounted cash flow method, binomial «decision tree» model and Black–Scholes model was carried out. The results have shown the advantage of option-based approach to the evaluation of IT project efficiency in comparison with the traditional DCF method, which allows to build flexibility in the planning and management of the project, assess its potential and consider the uncertainties as additional opportunities for profit.

Associated Factors with the Mortality Rate in Patients with COVID-19 - Decision Trees Vs. Logistic Regression

BACKGROUND Given the global burden of COVID-19 mortality, this study intended to determine the factors affecting mortality in patients with COVID-19 using decision tree analysis and logistic regression model in Kermanshah province, 2020. METHODS This cross-sectional study was conducted on 7799 patients with COVID-19 admitted to the hospitals of Kermanshah province. Data gathered from February 18 to July 9, 2020, were obtained from the vice-chancellor for the health of Kermanshah University of Medical Sciences. The performance of the models was compared according to the sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve. RESULTS According to the decision tree model, the most important risk factors for death due to COVID-19 were age, body temperature, admission to intensive care unit (ICU), prior hospital visit within the last 14 days, and cardiovascular disease. Also, the multivariate logistic regression model showed that the variables of age [OR = 4.47, 95 % CI: (3.16 -6.32)], shortness of breath [OR = 1.42, 95 % CI: (1.0-2.01)], ICU admission [OR = 3.75, 95 % CI: (2.47-5.68)], abnormal chest X-ray [OR = 1.93, 95 % CI: (1.06-3.41)], liver disease [OR = 5.05, 95 % CI (1.020-25.2)], body temperature [OR = 4.93, 95 % CI: (2.17-6.25)], and cardiovascular disease [OR = 2.15, 95 % CI: (1.27-3.06)] were significantly associated with the higher mortality of patients with COVID-19. The area under the ROC curve for the decision tree model and logistic regression was 0.77 and 0.75, respectively. CONCLUSIONS Identifying risk factors for mortality in patients with COVID-19 can provide more effective interventions in the early stages of treatment and improve the medical approaches provided by the medical staff. KEY WORDS COVID-19, Decision Tree, Logistic Regression, Mortality, Risk Factor

Understanding the Effect of Energy Density and Formulation Factors on the Printability and Characteristics of SLS Irbesartan Tablets—Application of the Decision Tree Model

Selective laser sintering (SLS) is a rapid prototyping technique for the production of three-dimensional objects through selectively sintering powder-based layer materials. The aim of the study was to investigate the effect of energy density (ED) and formulation factors on the printability and characteristics of SLS irbesartan tablets. The correlation between formulation factors, ED, and printability was obtained using a decision tree model with an accuracy of 80%. FT-IR results revealed that there was no interaction between irbesartan and the applied excipients. DSC results indicated that irbesartan was present in an amorphous form in printed tablets. ED had a significant influence on tablets’ physical, mechanical, and morphological characteristics. Adding lactose monohydrate enabled faster drug release while reducing the possibility for printing with different laser speeds. However, formulations with crospovidone were printable with a wider range of laser speeds. The adjustment of formulation and process parameters enabled the production of SLS tablets with hydroxypropyl methylcellulose with complete release in less than 30 min. The results suggest that a decision tree could be a useful tool for predicting the printability of pharmaceutical formulations. Tailoring the characteristics of SLS irbesartan tablets by ED is possible; however, it needs to be governed by the composition of the whole formulation.

Simplified Decision-Tree Algorithm to Predict Falls for Community-Dwelling Older Adults

The present study developed a simplified decision-tree algorithm for fall prediction with easily measurable predictors using data from a longitudinal cohort study: 2520 community-dwelling older adults aged 65 years or older participated. Fall history, age, sex, fear of falling, prescribed medication, knee osteoarthritis, lower limb pain, gait speed, and timed up and go test were assessed in the baseline survey as fall predictors. Moreover, recent falls were assessed in the follow-up survey. We created a fall-prediction algorithm using decision-tree analysis (C5.0) that included 14 nodes with six predictors, and the model could stratify the probabilities of fall incidence ranging from 30.4% to 71.9%. Additionally, the decision-tree model outperformed a logistic regression model with respect to the area under the curve (0.70 vs. 0.64), accuracy (0.65 vs. 0.62), sensitivity (0.62 vs. 0.50), positive predictive value (0.66 vs. 0.65), and negative predictive value (0.64 vs. 0.59). Our decision-tree model consists of common and easily measurable fall predictors, and its white-box algorithm can explain the reasons for risk stratification; therefore, it can be implemented in clinical practices. Our findings provide useful information for the early screening of fall risk and the promotion of timely strategies for fall prevention in community and clinical settings.

Identification of Neuromuscular Performance Parameters as Risk Factors of Non-contact Injuries in Male Elite Youth Soccer Players: A Preliminary Study on 62 Players With 25 Non-contact Injuries

Introduction: Elite youth soccer players suffer increasing numbers of injuries owing to constantly increasing physical demands. Deficits in neuromuscular performance may increase the risk of injury. Injury risk factors need to be identified and practical cut-off scores defined. Therefore, the purpose of the study was to assess neuromuscular performance parameters within a laboratory-based injury risk screening, to investigate their association with the risk of non-contact lower extremity injuries in elite youth soccer players, and to provide practice-relevant cut-off scores.Methods: Sixty-two elite youth soccer players (age: 17.2 ± 1.1 years) performed unilateral postural control exercises in different conditions, isokinetic tests of concentric and eccentric knee extension and knee flexion (60°/s), isometric tests of hip adduction and abduction, and isometric tests of trunk flexion, extension, lateral flexion and transversal rotation during the preseason period. Non-contact lower extremities injuries were documented throughout 10 months. Risk profiling was assessed using a multivariate approach utilizing a Decision Tree model [Classification and Regression Tree (CART) method].Results: Twenty-five non-contact injuries were registered. The Decision Tree model selected the COP sway, the peak torque for knee flexion concentric, the functional knee ratio and the path of the platform in that hierarchical order as important neuromuscular performance parameters to discriminate between injured and non-injured players. The classification showed a sensitivity of 0.73 and a specificity of 0.91. The relative risk was calculated at 4.2, meaning that the risk of suffering an injury is four times greater for a player, who has been classified as injured by the Decision Tree model.Conclusion: Measuring static postural control, postural control under unstable condition and the strength of the thigh seem to enable a good indication of injury risk in elite youth soccer players. However, this finding has to be taken with caution due to a small number of injury cases. Nonetheless, these preliminary results may have practical implications for future directions in injury risk screening and in planning and developing customized training programs to counteract intrinsic injury risk factors in elite youth soccer players.