An overview of the system.
The system is a web portal where the user has a unique username for handling their own data. Figure 1 shows the front page when the user logged in. For easy handling, the examinations are separated into modalities; computed tomography, conventional x-ray, interventional radiology, dentist CBCT and nuclear medicine.
For children a number of exams are included in the system. In total 6 types of examinations for computed tomography, 6 conventional x-ray exams and 3 interventional procedures. It should be emphasized that it could be difficult to establish diagnostic reference levels for all these exams. Figure 2 gives an example for the CT modality and the division into different exam groups based on patient age group and clinical indication.
Protocol parameters regarding the exam are included.
A number of technical parameters from the clinical protocol are reported and stored in the database. An example of such data is given in Figure 3.
Dose data, weight, height, age and sex are stored for each patient study. The local diagnostic reference value (LDRL), the median values, is calculated in the system. The patient's weight is also collected because in the future we intend to derive NDRL curves based on weight. Figure 4 shows the data format when collecting exam specific data.
Quality control is built-in the system.
Data is displayed to the user before sending data to the database. This has been introduced as a quality control mechanism. Radiation dose values are shown as a function of weight or BMI. An exception is examinations of the head where the valuations are shown consecutively. Outliers in the data set can then be identified. Figure 5 shows this for a group of patients.
All data in the database are available to users. A histogram summarizes data for a study type. The histogram of the NDRL for CT exam, head trauma is shown in Figure 6. The histograms allow the user to compare measured values with other hospitals.
Detailed information about the hospitals included in the histogram is also available and an example is shown in Figure 7. This information can be combined with the data of the protocol from each hospital, shown in figure 3.