<title>Comparison of hard and soft copy viewing of computed radiography portable chest radiographs</title>

Soft-copy display is emerging as a practical means of efficiently interpreting portable computed radiography exams. This study comparing the soft and hard-copy presentation of portable chest radiographs was undertaken to evaluate radiologist preference and confidence prior to implementing filmless soft-copy reading of portable chest radiographs. Seven radiologists with substantial previous experience interpreting portable chest radiographs directly compared 126 hard- and soft-copy presentations of computed radiography chest radiographs obtained over a two week period from a cardiac care unit. The radiologists first viewed the soft-copy on 1k X 1k monitors, using magnification and windowing tools when desired. Immediately following interpretation, a hard-copy produced from the same 2k data set and processed with identical parameters was reviewed and subjective comparisons recorded regarding the visibility and definition of lung pathology, soft tissues, bone detail and catheters. The use of magnification and windowing tools was recorded. The resulting data showed little perceived difference between the hard- and soft-copy images. The unmodified soft-copy images were considered equivalent for diagnostic purposes in regards to lung pathology in 78 percent, soft tissue 84 percent, bone detail 68 percent and catheters 76 percent. The hard-copy was more frequently considered better for bone detail and catheter visualization. The soft-copy was more frequently considered better for visualizing lung pathology and soft tissue structures. Changing the window settings and magnification comparison of images eased their concerns about the adequacy of soft-copy presentation of computed radiography chest images at least the equivalent of hard-copy for depiction of normal anatomy and pathologic features in most categories. The next step is to optimize the soft-copy display parameters for routine viewing and then for specific clinical questions.