Search

The effects of misregistration artifacts and background corrections on the densitometric measurement of left ventricular ejection fraction (EF) from digital subtraction angiography (DSA) images were studied in 20 patients. Densitometric ejection fraction measurements were performed on both conventional time subtraction images and on dual-energy subtraction images. Dual-energy subtraction is not sensitive to the motion induced artifacts which often mar time subtraction images. While the time subtraction images had varying degrees of misregistration artifacts, none of the dual-energy studies contained significant misregistration artifacts. Densitometrically determined ejection fractions measured with and without correction for background signals were compared. Poor agreement between time subtraction ejection fractions determined with and without background correction (EFNO-BKG = 0.88 EFBKG - 6.0%, SEE = 8.1%, r = 0.83) demonstrated the sensitivity of time subtraction EFs to the performance of a background correction procedure. Conversely, densitometric measurement of ejection fraction using dual-energy subtraction was significantly less sensitive to the performance of a background correction (EFNO-BKG = 0.99 EFBKG - 5.3%, SEE = 4.3%, r = 0.96). Since background correction requires accurate definition of ventricular borders, but motion artifacts often preclude accurate border definition, it is concluded that dual-energy subtraction is a significantly more robust method for measuring left ventricular ejection fraction using densitometry. It is further concluded that identification of the systolic endocardial border is not required when performing densitometric EF measurements on dual-energy images. Drawing of the end-diastolic border alone is sufficient to produce an accurate ejection fraction measurement.

Comparative calibration is the broad statistical methodology used to assess the calibration of a set of p instruments, each designed to measure the same characteristic, on a common group of individuals. Different from the usual calibration problem, the true underlying quantity measured is unobservable. Many authors have shown that this problem, in general, does not have a unique solution. Most commonly used assumptions to obtain a unique solution are (i) one instrument is the gold standard (that is, unbiased) and (ii) the measurement errors of the p instruments are independent. Such constraints, however, may not be valid for many clinical applications, for example, the universal standardization project for dual X-ray absorptiometry (DXA) scanners. In this paper, we propose a new approach to resolve the comparative calibration problem when a gold standard is unavailable. Instead of the usual assumptions, we use external information in addition to data from the p instruments, to solve the problem. We address statistical estimation, hypothesis testing and missing data problems. We apply the new method specifically to the universal standardization project data where a group of individuals have been measured for bone mineral density (BMD) by three DXA scanners. We compare the results of the new method to currently used methods and show that they have better statistical properties.