BackgroundRespiratory rate (RR) is an important vital sign used in the initial and ongoing assessment of unwell children. It is used in Paediatric Early Warning Scores to assess a child’s clinical ...status and as a predictor of serious deterioration. Convenient electronic devices exist for the measurement of pulse, blood pressure, oxygen saturation and temperature that provide accurate and prompt measures of vital signs. Devices for monitoring RR have entered the commercial market but there is no device currently available that gives an accurate and rapid assessment of RR in clinical practice.AimWe have developed a contactless portable respiratory rate monitor (CPRM) and aimed to measure agreement with existing methods of RR measurement.MethodRespiratory rate data were collected from 30 children undergoing polysomnography sleep studies at a tertiary children’s hospital. Measurements from respiratory inductance plethysmography (RIP) bands (established contact method), visual counting of chest movements (established non-contact method) and the CPRM (developed device) were all obtained simultaneously and compared. Two to three data sets were collected for each child. A total of 61 recordings were obtained from children ranging from 8 months to 15 years.ResultsData showed substantial agreement between measurements from the CPRM and the gold standard RIP (Interclass correlation coefficient 0.762: 95% CI 0.633–0.850). When two patients with significantly dysfunctional breathing were removed from the analysis, the CPRM showed a much higher correlation with the gold standard RIP method (Interclass correlation coefficient 0.981: 95% CI 0.968- 0.989) with 95% limits of agreement –2.49–0.77 breaths/min (Figure 1).Abstract G490 Figure 1 Bland-Altman plot assessing pairwise agreement between CPRM and RIPConclusionA portable contactless device developed in this study can accurately and quickly measure respiratory rate. Such a device will be an important tool in the initial assessment of unwell children. There is currently no such device on the market. More testing is needed to explore outlying measurements and to evaluate in different clinical settings. Further development and modification of the device and software are ongoing.
The most common measure of association between two continuous variables is the Pearson correlation (Maronna et al. in Safari an OMC. Robust statistics, 2019. ...https://login.proxy.bib.uottawa.ca/login?url=https://learning.oreilly.com/library/view/-/9781119214687/?ar&orpq&email=^u). When outliers are present, Pearson does not accurately measure association and robust measures are needed. This article introduces three new robust measures of correlation: Taba (T), TabWil (TW), and TabWil rank (TWR). The correlation estimators T and TW measure a linear association between two continuous or ordinal variables; whereas TWR measures a monotonic association. The robustness of these proposed measures in comparison with Pearson (P), Spearman (S), Quadrant (Q), Median (M), and Minimum Covariance Determinant (MCD) are examined through simulation. Taba distance is used to analyze genes, and statistical tests were used to identify those genes most significantly associated with Williams Syndrome (WS).
Based on the root mean square error (RMSE) and bias, the three proposed correlation measures are highly competitive when compared to classical measures such as P and S as well as robust measures such as Q, M, and MCD. Our findings indicate TBL2 was the most significant gene among patients diagnosed with WS and had the most significant reduction in gene expression level when compared with control (P value = 6.37E-05).
Overall, when the distribution is bivariate Log-Normal or bivariate Weibull, TWR performs best in terms of bias and T performs best with respect to RMSE. Under the Normal distribution, MCD performs well with respect to bias and RMSE; but TW, TWR, T, S, and P correlations were in close proximity. The identification of TBL2 may serve as a diagnostic tool for WS patients. A Taba R package has been developed and is available for use to perform all necessary computations for the proposed methods.
•Three correlation coefficients of hydration heat and strength are compared.•Linear and nonparametric correlation analysis is used quantitatively.•More variables are considered, the greater of the ...multiple correlation coefficient.
Analysis of linear and nonlinear dependencies of hydration characteristics and strength development is of interest for reliable and cost-effective designs. In this paper, the statistical assessment of linear and nonparametric correlation analysis model is used to investigate the association of these properties quantitatively. The bivariate correlation of early hydration characteristics within 72 h and compressive strength at different curing ages is evaluated by Pearson’s, Spearman’s rho and Kendall’s tau correlation coefficient, respectively. Assessment results of various methods show that early hydration characteristics and compressive strength have a strong correlation coefficient. Furthermore, the coefficient fitted by Spearman’s rho correlation analysis is higher than those by Pearson’s and Kendall’s tau analyses. The calculated correlation coefficients of middle and long curing ages (e.g. 28 d and 56 d) are higher than those in the early curing ages (e.g. 3 d and 7 d). For multiple correlation analysis, the correlation coefficients between heat release characteristics and mechanical properties undergo a fundamental change. The more variables of hydration characteristic are considered, the greater the multiple correlation coefficient of compressive strength, and the coefficients of middle and long curing age strength have a narrower range than early age compressive strength. Hence, the linear and nonparametric correlation model is a useful quantitative evaluation method for assessing the relationship between the early hydration characteristics and compressive strength for multi-composite blends.
Knowledge of spatial correlations of precipitation is important for the generation of grid‐based surface precipitation data sets, deployment of data collection, selection of downscaling strategies, ...and interpretation of paleoclimate reconstructions. Spatial correlations of daily precipitation in China were analyzed based on a daily precipitation data set from 1951 through 2014 for 2,208 stations by dividing them into 13 regions. Interstation Pearson correlation coefficient r for the daily precipitation series and the corresponding interstation distance d were calculated for each region. The exponential spatial correlation model (rd=c0×exp−d/d0s0+1−c0) was fitted by the r‐d pairs, in which c0, d0 and s0 were the parameter variance, scale and shape, respectively. The results showed that: (a) The determination coefficient R2 of the correlation model varied from 0.54 to 0.96, with a mean of 0.82 and the regional maximum correlation distance d0 varied from 102.2 to 201.7 km, with a mean of 155.2 km. Western regions generally had smaller d0 than eastern regions, which indicates rain events in the western regions were more local; (b) The goodness‐of‐fit of the model was improved by dividing samples into West‐East (W‐E) and North‐South (N‐S) directions. The average of d0 for all regions (190.2 km) for the W‐E direction is larger than that for the N‐S direction (142.9 km); (c) The correlation distances in summer and dry years are shorter than those in winter and wet years. However, the difference of correlation distance between dry and wet years was subtle compared with those between summer and winter, and between W‐E and N‐S directions. Seven regions were divided based on the spatial correlations of daily precipitation and different spatial models were suggested to be used for different regions, seasons and directions when the interpolation of daily precipitation is conducted for the generation of gridded surface precipitation data sets.
Spatial correlations of daily precipitation in China were analyzed by fitting exponential spatial correlation models between the interstation Pearson correlation coefficient for daily precipitation series and the corresponding interstation distance based on the daily precipitation data from 1951 through 2014 for 2,208 stations. Seven regions were divided based on the spatial correlations of daily precipitation and different spatial correlation models were suggested to be used for different regions, seasons and directions when the interpolation of daily precipitation is conducted.
In equation (1a) in function f(), the subscript for the first term within the brackets should be 1k,t, not ik,t. Please view the correct equation here: A general requirement for the stability of this ...equilibrium is thatd/e > a(R0 – K)/(R0 + K)" Citation: Ruokolainen L (2013) Correction: Spatio-Temporal Environmental Correlation and Population Variability in Simple Metacommunities.
Aims This study looked at aspects of the directed cord blood programme at a paediatric centre. The aims of the study were to:- Assess the percentage of directed cord blood units which are HLA ...matched. Assess the adequacy of total nucleated cell (TNC) dose and CD34+ cell dose from the collected cord blood based on recommendations from the European school of Haematology contained in the EBMT handbook of 2012. Establish if CD34+ and TNC count correlates with the volume of cord blood collected. Methods Seventeen consecutive cord blood collections were identified from October 2008 to July 2013. Information from patient records was collated onto a spread sheet. Cell counts were converted into dose per kilogram using the recipient’s weight or 50th centile weight for the recipient’s age and gender where weight was unavailable. The EBMT handbook recommends the following cell doses: TNC: 2.5–3 × 107/kg CD34+: 1.2–1.7 × 105/kg Results There was a higher than expected level (35% vs. 25%) of HLA matching between the siblings in the study. 58.8% met the recommended guidelines for TNC dose and 41.2% for CD34+ cell dose – see Figures 1 and 2. Figure 1 Abstract G252(P) Figure 1 Figure 2 Abstract G252(P) Figure 2 Both adequate TNC and CD34+ cell doses were achieved in 41.2% of cases. This study showed a positive correlation between volume of cord blood and TNC count (R2 value 0.85), the relationship between CD34+ count and volume is less strongly correlated (R2 value 0.53) – see Figure 3. Figure 3 Abstract G252(P) Figure 3 Conclusion More than one third of collections were HLA identical to the intended recipient. Over 50% achieved an adequate cell dose for haematopoietic stem cell transplantation. There is a strong correlation between volume of cord blood collected and TNC count, but not for CD34+ count in this sample. Adequacy of cord blood is judged on TNC count and is assumed to reflect the CD34+ component, which is essential for haematopoietic engraftment. All cord blood collected in this series came from parous mothers following second or subsequent births. These factors together with maternal age are known to affect cord blood cell count.
To evaluate binary classifications and their confusion matrices, scientific researchers can employ several statistical rates, accordingly to the goal of the experiment they are investigating. Despite ...being a crucial issue in machine learning, no widespread consensus has been reached on a unified elective chosen measure yet. Accuracy and F
score computed on confusion matrices have been (and still are) among the most popular adopted metrics in binary classification tasks. However, these statistical measures can dangerously show overoptimistic inflated results, especially on imbalanced datasets.
The Matthews correlation coefficient (MCC), instead, is a more reliable statistical rate which produces a high score only if the prediction obtained good results in all of the four confusion matrix categories (true positives, false negatives, true negatives, and false positives), proportionally both to the size of positive elements and the size of negative elements in the dataset.
In this article, we show how MCC produces a more informative and truthful score in evaluating binary classifications than accuracy and F
score, by first explaining the mathematical properties, and then the asset of MCC in six synthetic use cases and in a real genomics scenario. We believe that the Matthews correlation coefficient should be preferred to accuracy and F
score in evaluating binary classification tasks by all scientific communities.
PDF
