 Lost Souls
|
 Peds Atlas
|
 Volumes
|
 Aortic Diameter - Adult
|
 Kidney Size - Peds
|
 Spleen Size - Peds
|
 Testicular Volume
|
 Lung Nodule 2017
|
 Contact
|
Aortic and Pulmonary Artery Diameter Percentile Calculator (Adult)
contributed by Michal Kulon, MD on 9/15/2015
| Male | Diameter(mm) | Percentile | Z-score | Group Age Range | Group BSA Range | Average | Standard Deviation |
| Ascending | 0 | 0 | -13 | 0 - 45 | 0 - 1.7 | 28.6 | 2.2 |
| Descending | 0 | NA | NA | 0 - 45 | 0 - 1.7 | 20.9 | NA |
| Pulmonary | 0 | Observed:0 Gaussian:0 |
-5.82 | 15.7 | 2.7 |
Methods
- Aorta Diameter Normal Range Data
- Data based on: Wolak A, Gransar H, Thomson LJ, et al. Aortic Size Assessment by Noncontrast Cardiac Computed Tomography: Normal Limits by Age, Gender, and Body Surface Area. J Am Coll Cardiol Img. 2008;1(2):200-209.
- Raw data was not published; the normality of the sizes within the raw data therefore could not be verified. Calculation of percentiles utilizes the published averages and standard deviations for the binned age and BSA groups and assumes a normal distribution of size diameters within each interval.
- Pulmonary Artery Diameter Normal Range Data
- Data based on: Karazincir S. et al., "CT assessment of main pulmonary artery diameter," Diagnostic and Interventional Radiology 14(2), 72-74 (2008).
- Raw data was not published. Derivation from the graph published in the article (figure 2) was therefore necessary. This was done by applying a black flood-fill to the background of the graph image, and software implementation of Hough Transform, with the expectation of finding filled circles.
- Observed Method: Residuals of the study data were computed by subtracting the linear fit, and compared to the residual of the submitted diameter. The reported percentile reflects the percentage of the population with pulmonary artery size (diameter) less or equal to the submitted diameter.
- Gaussian Method (CAUTION): Regression residuals do not follow a normal distribution. Box-Cox transform of the data also does not follow a normal distribution; even for the optimally-transformed data (Box-Cox lambda=1.23), the Shapiro-Wilk test for normality rejects the hypothesis that the data derives from a normal distribution (P=0.0171). Nonetheless, the bounds of 2 standard deviations are numerically close to the largest of the residuals, and may therefore be approximately approriate as upper bound of size.
- Density and QQ plots of raw data, and QQ plot of the Box-Cox transformed data