Stereological estimation of left-ventricular volumetric and functional parameters from multidetector-row computed tomography data

Unbiased and efficient estimation of the volume of the fibroid uterus using the Cavalieri method and magnetic resonance imaging

The aim of our study was to develop a reliable technique for measuring volume of the fibroid uterus using Magnetic Resonance Imaging. We applied the Cavalieri method and standard calliper technique to measure the volume of the uterus and largest fibroid in 26 patients, and results were compared with "gold-standard" planimetry measurements. We found Cavalieri measurements to be unbiased, while calliper measurements systematically underestimated uterine volume (- 13.2%, P < 10(-5)) and had greater variance. Repeatability was similar for the 2 techniques (standard deviation [SD] = 4.0%-6.9%). Reproducibility of Cavalieri measurements was higher for measurement of uterine (SD = 9.0%) than fibroid volume (SD = 19.1%), whereas the reproducibility of calliper measurements was higher for fibroid (SD = 9.1%) than uterine volume (SD = 15.9%). The additional measurement time for the Cavalieri method was approximately 1 to 2 minutes. In conclusion, the Cavalieri method permits more accurate measurement of uterine and fibroid volumes and is suitable for application in both clinical practice and scientific research.

Assessment of left ventricular function and mass by MR imaging: a stereological study based on the systematic slice sampling procedure

RATIONALE AND OBJECTIVES

The aim of this study was to combine the stereological technique with magnetic resonance (MR) imaging data for the volumetric and functional analysis of the left ventricle (LV).

MATERIALS AND METHODS

Cardiac MR examinations were performed in 13 consecutive subjects with known or suspected coronary artery disease. The end-diastolic volume (EDV), end-systolic volume, ejection fraction (EF), and mass were estimated by stereology using the entire slice set depicting LV and systematic sampling intensities of 1/2 and 1/3 that provided samples with every second and third slice, respectively. The repeatability of stereology was evaluated. Stereological assessments were compared with the reference values derived by manually tracing the endocardial and epicardial contours on MR images.

RESULTS

Stereological EDV and EF estimations obtained by the 1/3 systematic sampling scheme were significantly different from those by manual delineation (P < .05). No difference was observed between the reference values and the LV parameters estimated by the entire slice set or a sampling intensity of 1/2 (P > .05). For these stereological approaches, a high correlation (r(2) = 0.80-0.93) and clinically acceptable limits of agreement were found with the reference method. Stereological estimations obtained by both sample sizes presented comparable coefficient of variation values of 2.9-5.8%. The mean time for stereological measurements on the entire slice set was 3.4 ± 0.6 minutes and it was reduced to 2.5 ± 0.5 minutes with the 1/2 systematic sampling scheme.

CONCLUSION

Stereological analysis on systematic samples of MR slices generated by the 1/2 sampling intensity provided efficient and quick assessment of LV volumes, function, and mass.