Lucia J.M. Kroft MD, Presenter: Nothing to Disclose

Wouter J.H. Veldkamp PhD, Abstract Co-Author: Nothing to Disclose

Bart J.A. Mertens PhD, Abstract Co-Author: Nothing to Disclose

Jan-Pieter A. van Delft BS, Abstract Co-Author: Nothing to Disclose

Jacob Geleijns PhD, Abstract Co-Author: Nothing to Disclose

To determine if dose reduction results in decreased nodule detection for digital chest radiography.

Raw data of digital PA chest images obtained with a slot-scan CCD chest radiography system in 20 patients reported with a normal chest examination were used. One hundred and twenty lesions varying in size and density, and representing nodules were digitally simulated and added to these radiographs. The lungs and mediastinum were separately investigated. Hardcopies were printed with 100% dose and, by adding noise, with simulated patient doses of 50%, 25%, and 12%. Four radiologists performed reading. Each added lesion was assigned one of two outcome scores: “detected” or “not detected”. False positive readings were also scored. A semi-parametric logistic regression model with correction for patient and observer effect was used for statistical analysis.

For the lungs, no effect was found for lesion detection with decreased radiation dose from 100% to 50%, 25%, or 12%. However, for the mediastinum lesion detection probabilities deteriorated from the 100% dose to the 50%, 25%, and 12% dose with probabilities of 0.97 (95% C.I. -0.86, 0.012), 1 (C.I. -0.59, -0.61), and 1 (C.I. -2.41, -1.22), respectively, with the C.I.’s for the effects on the log(odds). For the lungs as well as for the mediastinum, no differences were found in the numbers of false positive observations between the doses.

For digital radiography, dose reduction resulted in decreased detection of simulated nodules in patients for the mediastinum but not for the lungs, whereas dose reduction had no effect on false positive observations. Decreasing dose for digital chest radiography may not be advisable when sought for nodules, with detection in the mediastinum being a limiting factor.

Kroft, L, Veldkamp, W, Mertens, B, van Delft, J, Geleijns, J, Simulated Lesion Detection for Reduced Dose Digital Chest Radiography: Effect on Lesion Detection in the Mediastinum But Not in the Lungs.  Radiological Society of North America 2005 Scientific Assembly and Annual Meeting, November 27 - December 2, 2005 ,Chicago IL. http://archive.rsna.org/2005/4407071.html