*4.5. Statistics*

Statistical analyses were performed by using SPSS version 25 (IBM, Armonk, NY, USA) and Prism 7.0 software (GraphPad Software Inc., San Diego, CA, USA). ROUT method was performed for outlier identification with an average false discovery rate less than 1%. After testing for normal distribution group differences were analyzed by 2-tailed unpaired t-test or by Mann–Whitney U test. All data are presented as means ± SEM. To describe the quality and the sensitivity of the STE measurements in both species, intra- and interobserver comparisons were presented for EF and GLS. Intra- and interobserver variability was assessed on all animals and all study participants by the repeated evaluation of the EF and GLS. Pearson correlation, Bland–Altman plots, and intraclass correlation were performed to assess the agreemen<sup>t</sup> and the reliability within and between the observers. For interobserver comparison, two experts evaluated parameters of EF and GLS independently in a blinded manner concerning the animal model as well as in the human participants. For intraobserver comparison, one observer repeatedly evaluated EF and GLS in a blinded fashion. Bland–Altman plots were reported including the mean difference (bias) with corresponding 95% limits of agreemen<sup>t</sup> between the evaluations of the two observers (interobserver agreement) or the repeated evaluations of one observer (intraobserver

agreement). For the calculation of the intraclass correlation coe fficient, a two-way random-e ffect model based on single measurements and absolute agreemen<sup>t</sup> assessed the interobserver repeatability, and a two-way mixed-e ffect model based on single measurements assessed the intraobserver repeatability for one observer. Mean estimators with 95% confidence intervals (CI) were reported for each ICC. Interpretation of ICC and Pearson correlation coe fficients was elaborated as the following: <0.50 = poor; between 0.50 and 0.75 = moderate; between 0.75 and 0.90 = strong; >0.90 = excellent. A two-sided *p* < 0.05 was considered statistically significant.
