Development And Application Of A Java Program For Extracting Signal Data From DICOM Files

ObjectiveTo develop a Java software program for extracting and quantitatively analyzing the signal data from DICOM imaging files, and compare the results with those of commercialized software.Materials and methodsUncompressed DICOM files were copied from the Siemens Sensation Series multi-detector-row (4-slice and 16-slice) spiral CT machines, and the formats of these files were analyzed. According to the result of the analysis, a Java application software program was developed to extract the imaging signal data from the files. Based on the selected region of interest (ROI), the parameters of signal data within each ROI were calculated. Those parameters were used for quantitative analysis according to different clinic requirements, and the results were compared with those being measured with commercial software installed on Siemens CT.Result1> The software program successfully extracted the signal data from DICOM files and calculated various parameters (area, volume, count,total value, average value, standard deviation, maximum value and the minimum value) of each selected ROI.2> Some simple clinic quantitative analyses were tested using this software program, such as the measurement of the area of abdominal fat, drawing of time-density curve and coronary artery calcium scoring. The results showed very close consistence with those measured with commercial software installed on Siemens CT.ConclusionA Java software program was successfully developed for extracting and quantitatively analyzing the signal data from DICOM imaging files. The test results of some simple clinic quantitative analyses were closely comparable to those measured by commercial software.