Related Experiment Video
Updated: Aug 20, 2026
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
Soft copy using image processing in place of hard copy for detection of subtle pulmonary lesions: is it actually
Katsumi Abe1, Shigeru Kosuda, Yoshie Iwasaki
1Received March 4, 2004; revision accepted April 20, 2004. Department of Radiology, National Defense Medical College, Saitama, Japan.
Purpose:
To comparatively assess the interpretation performance of hard-copy versus soft-copy presentations for detecting subtle pulmonary lesions and to estimate soft-copy cost-effectiveness.
Methods:
Computed radiography was used to obtain images with a 1,600x 1,200 matrix having 8 bits of gray-scale definition. The two separate image formats (hard and soft copy), which consisted of 60 chest radiograph combinations, were shown to five board-certified radiologists. The costs of hard copies, chemicals, waste disposal, and labor were calculated.
Results:
The mean accuracies and Az values were 0.63 (hard) vs. 0.64 (soft) and 0.657 (hard) vs. 0.729 (soft), respectively. Reading one set of hard copies took on average 8.0 sec less than reading soft copies (30.3 vs. 38.3 sec). Estimated savings resulting from replacement of hard copy with soft were US$128,004 per year or US$6.20 per patient at our hospital.
Conclusion:
Conversion from hard copy to soft copy, using image processing, is feasible and cost-effective.
