LCD bests CRT in observer performance study

A study using both human and model observers to compare the performance of LCD and CRT monitors gives the edge to LCD displays, affirming a shift in the marketplace to the flat-panel systems.

"A critical issue today is whether flat-panel liquid crystal displays are more suitable for primary diagnostic reading than the more traditional cathode ray tube," said Elizabeth A. Krupinski, Ph.D., a research professor in radiology at the University of Arizona.

Krupinski's project evaluated a human visual system model (JNDmetrix) based on just noticeable difference (JND) and frequency-channel vision-modeling principles to assess whether a CRT or an LCD monochrome display monitor would yield better observer performance in radiographic interpretation.

The results demonstrate that an LCD monitor with lower veiling glare and isotropic modulation transfer function (MTF) offers significantly better performance than a CRT monitor with higher veiling glare and nonisotropic MTF, Krupinski said in the study (J Digit Imaging, epub 24 August 2004).

The results apply at all lesion contrast levels, although it was only at the 75% and 50% contrast levels that statistical significance was reached for the human observers, according to the paper.

"The human and model results closely parallel one another and demonstrate essentially the same pattern of results," Krupinski said. "In terms of the model, this provides us with further evidence that the JNDmetrix model is a suitable tool for helping to predict observer performance in radiologic image interpretation tasks as a function of the manipulation of various viewing parameters."

Krupinski offered a few caveats, however.

"The study used only mammographic images with masses, so it is not possible to generalize completely to other types of images and lesions," she said.

It is also important to remember that the study used only small regions of interest from the original mammograms, which eliminated the need for the observers to engage in extensive searching.

"It is difficult to say whether performance would be better or worse if we had used the full images and the observers had to search for the lesions," Krupinski said.

It is highly likely, however, that although overall performance would likely decrease (since readers would have to search the entire image and detection errors would therefore be more likely), the same general pattern of results would be observed, she said.