Format

Send to

Choose Destination
Clin Nucl Med. 2012 Apr;37(4):356-61. doi: 10.1097/RLU.0b013e3182443f8c.

Automatic measurement of renal volume in children using 99mTc dimercaptosuccinic acid SPECT: normal ranges with body weight.

Author information

1
Division of Nuclear Medicine and Molecular Imaging, Children's Hospital Boston, Harvard Medical School, Boston, MA 02115, USA.

Abstract

OBJECTIVE:

The purpose of this study was to automate volume measurement of pediatric kidneys using Technetium-99m-dimercaptosuccinic acid (99mTc DMSA) SPECT. Assessment of renal size is essential in making an accurate clinical diagnosis and in evaluating normal and abnormal renal growth. Existing techniques for measuring renal length and volume are often observer-dependent and are limited by interobserver variability.

MATERIALS AND METHODS:

An automated method using 99mTc DMSA SPECT was developed to measure renal length, volume, and 2D/3D split renal functions. An adaptive threshold was used for segmentation of renal images. To improve measurement accuracy of renal volumes, a boundary-weighted integration and a linear correction from phantom experiments were applied. Using the new automated software, we evaluated 620 normal kidneys in 310 patients aged 1 to 21 years.

RESULTS:

In all, 99.4% of renal volume measurement was inter-/intraobserver reproducible without manual intervention. Correlations between renal size (volume and length) and growth parameters (age and body weight) were evaluated by nonlinear regression modeling. Body weight was found to be the best predictor of kidney volume and length (P < 0.0001). Normal ranges were established according to body weight.

CONCLUSIONS:

Automatic renal volume measurement with 99mTc DMSA SPECT demonstrates high observer reproducibility, and it is useful for achieving accurate assessment of renal growth using sequential measurements. The renal growth models generated by this study will aid physicians in evaluating and monitoring renal growth.

PMID:
22391704
DOI:
10.1097/RLU.0b013e3182443f8c
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wolters Kluwer
Loading ...
Support Center