Iterative reconstruction algorithm aids lung CAD

Sunday, December 1 | 11:45 a.m.-11:55 a.m. | SSA19-07 | Room S403B
Japanese researchers will discuss in this Sunday session how an iterative reconstruction technique can improve the performance of computer-aided detection (CAD) for finding lung nodules across a range of different CT dose levels.

As part of a preliminary evaluation of a CT lung nodule CAD system developed by Kobe University Graduate School of Medicine in collaboration with Toshiba Medical Systems, researchers explored the effect of Toshiba's adaptive iterative dose reduction (AIDR) 3D technology on the standalone performance of the CAD prototype using CT data acquired at standard-dose, low-dose, and ultralow-dose levels.

They found that CAD with AIDR 3D on standard-dose CT surpassed the performance of standard-dose CT alone. With low-dose and ultralow-dose studies, CAD with AIDR 3D turned in similar results to standard-dose exams, according to the researchers.

"It is tempting to think that the performances of CAD systems for lung nodules at large can be improved by using the latest reconstruction algorithms such as AIDR 3D," presenter Dr. Sumiaki Matsumoto, PhD, told AuntMinnie.com. "However, it is not clear whether the findings of this study [can be generalized to other systems]."

Page 1 of 654
Next Page