[Automated Detection of Radiopaque Markers in Chest and Abdomen Radiography].

We aimed to develop a computerized method for the detection of radiopaque markers, such as R and L in chest and abdomen radiography by using the generalized Hough transform and the template matching. To develop the computerized method, we used 200 chest and abdomen images in our institution as training cases. First, two template images for R and L markers were created with the same exposure condition as a chest X-ray. Following various image processing, such as edge detection, thinning and Hough transformed, a look-up table that consisted of distance and direction pairs was built for the generalized Hough transform. All training images were preprocessed with median filter, edge detection, binarization, thinning, back ground removal and labeling. For candidates of markers that were detected as true positive or false positive, their vote and cross-correlation were calculated with the generalized Hough transform. To evaluate this proposed method, a validation test was performed with another database that consisted of 800 chest and abdomen images by use of Mahalanobis distance based on vote and cross-correlation in statistics. The precision of detecting the radiopaque markers for 800 test images was 99.9%. In addition, this method worked out well for some specific images in which markers were overlapped with a human body.

[Development of software for the verification of patient flow through a daily clinical environment by use of the radiology information system (RIS)].

In order to manage relationship between patients' movements and operating efficiency, we developed a special software which can make patient flow visible on a display monitor by use of actual data obtained from the radiology information system (RIS). In this software, a simple floor map of the radiology department in our hospital was drawn on the monitor and each patient was indicated with a small figure. This software was developed with commercialized computer software [Excel 2007 visual basic applications (VBA) Microsoft]. Movements of the patient figures were simulated by use of actual time data such as registration of radiology department, and start and ending time of examinations. The patient figures were moved along with predetermined flow lines every second. The movements of the patient figures were controlled by several buttons (i.e., play and stop) and setting switches for determining reproduction date and time. In conclusion, by use of this software, the patient flows could be analyzed systematically by checking efficient operation such as average waiting time of the patients and/or standby time of radiological technologists.

Clinical utility of ultra-low-dose pre-test exposure to avoid unnecessary patient exposure due to positioning errors: a simulation study.

The use of digital radiographic systems has decreased the frequency of image retakes due to over/underexposure in general radiography. However, image retakes owing to patient positioning errors are likely to increase because of the convenience of a real-time image check on a console table. The purpose of the present study is to propose a novel radiographic examination procedure with an ultra-low-dose pre-test exposure that may be utilized to check patient positioning prior to taking an actual image, thereby reducing unnecessary patient exposure owing to image retakes. In this study, examination data from 714 knee joint radiographs, both submitted and retaken images, were included. Twelve radiological technologists (RTs) took all images. The actual total exposure dose for each patient was compared with simulated total doses utilized in the proposed procedure. The simulation assumed that each examination was completed following pre-test exposure. Therefore, this method did not involve retaking images although at least one pre-test exposure had been applied to all patients. Pre-test exposures at four dose levels corresponding to 25, 10, 5, and 2% of the actual exposure dose were evaluated to determine whether each dose level could be used to check patient positioning. The results indicated that when the pre-test exposure dose rate was 10% or lower, the total exposure dose reduction equaled or exceeded 8% for all patients. The use of the proposed procedure reduced the total exposure dose for all patients when compared to the exposure dose calculated from records.

A skyline-view imaging technique for axial projection of the patella: a clinical study.

Our purpose in this study was to evaluate the clinical usefulness of a new skyline-view imaging technique for axial projection of the patella with use of the anterior border of the patella and tibial tuberosity as position indicators. Our database consisted of pairs of axial images of the patella of the same patients, obtained with use of conventional and new techniques for the radiographic diagnosis of knee-joint diseases. A total of 118 pairs of knee images were obtained from 103 patients ranging in age from 16 to 86 years (mean age 49.7 years). The patellar axial positioning errors were determined in each of the images obtained with the two techniques. The relative error according to the patellar tilt was determined from each of the axial images of the patellas of the same patients obtained with the conventional and new techniques for the radiographic diagnosis of knee-joint diseases. The patellar axial positioning error was 0.40 with the conventional technique, whereas that with the new technique was significantly different at 0.30. This clinical study confirmed that the new skyline-view imaging technique, which uses the anterior border of the patella and the tibial tuberosity as position markers that can be confirmed by palpation, provides more accurate axial images than the conventional imaging technique.

Appropriate incidence angle for fundamental research on new skyline radiography development.

Our purpose in this study was to elucidate the relationship between the angles formed between the anterior patella, tibial tuberosity, and the knee joint cavity and the flexion angle, sex, and age of the subjects. We investigated 368 images of 280 patients ranging in age from 16 to 60 years (179 knees of 150 men, mean age 36.4 years; 189 knees of 130 women, mean age 41.4 years) who underwent lateral radiography of the knee. The tibial tuberosity on the lateral radiograph of the knee was defined as a reference point, and a line tangent to the anterior patella was used as a reference line. The angle between the reference line and the straight line from the reference point to the knee joint cavity (incidence angle) was measured. The average incidence angle was 19° (SD 1.8°). There was almost no correlation between the incidence angle and the flexion angle, and neither the knee flexion angle nor age had any influence on the incidence angle. There was a difference between the sexes in the average incidence angle, but this difference was less than 1°. Further study on the same patients is required for comparison of this technique with the conventional technique that uses the femur for reference.

A simple method for identifying image orientation of chest radiographs by use of the center of gravity of the image.

Bedside chest radiography is a frequent X-ray examination when patients are physically incapacitated. An X-ray cassette with an imaging plate is inserted below the patient's body, and the image orientation of the radiograph is determined by the direction of insertion. Therefore, an incorrect direction of insertion would yield an incorrect image orientation for diagnosis, if no correction was performed on the resulting image data. We aimed to develop a computerized method that identifies the image orientation of chest radiographs by using the center of gravity (COG) of the images in terms of pixel values. To develop the computerized method, we used 247 chest images contained in the Japanese Society of Radiological Technology database as training cases, and 1833 bedside chest radiographs obtained in our institution for validation testing. As a result for the 247 training images, the angles which were obtained from directions between the COG of pixel values and the center of the image were distributed between 162.7° and 224.4° in a clockwise direction. We used the angle of the COG to identify the correct view orientation. The range of angles (139.1°-229.0°) for the COG in the chest image with correct image orientation was determined with a 99 % confidence interval for the angles of the COGs obtained from the training images. As a result of the validation test based on the range of angles determined, 99.7 % of the 1833 test images were identified correctly.