Selecting one of the available sleep software will bring up the export selection window (Fig. 2). From here, the user can select the installation path of the sleep software, the path of the folder containing the recordings, and the path where the user wishes the exports to be saved. Each of these paths can be selected from a “browse for folder” menu. Alternatively, the user can directly select default paths, which can be changed in the settings menu. The user can also go back to the main selection window by clicking “Go back”.
The recordings should be placed in a single folder so that each subfolder in this folder contains the recording data for one measurement. An example of a correct folder structure is presented in Fig. 3. The user can also select which individual recordings should be exported from the recordings directory. Figure 4 shows the interface for selecting the individual recordings. The user can either individually check which recordings should be exported, check all recordings, or uncheck all recordings. Clicking “OK” will bring the user back to the export selection window.
The user can also select what data will be exported. This will vary slightly depending on the export options allowed by the corresponding sleep software. In the RemLogic example case, the options are EDF signal export, TXT event export, PSG report, XML event export and TXT sleep stage export. The user can tick one or multiple of these boxes and the corresponding files will be generated during the export process.
Finally, the user can select the general “speed” parameter. A standard computer that is capable of running Windows 10 and the sleep software in manual mode without issues can also easily handle the fastest export speed. However, some sleep centers may use dated hardware with limited capabilities for processing the sleep studies due to licensing or software support issues. In addition, the SSRE software may be used remotely where a bad connection might cause issues with the fastest speed. In these cases, slower export speeds can be selected for increased export time, but possibly higher compatibility. If a too fast speed is used on a slow computer, all of the simulated keystrokes may not be correctly registered, which might break the export as wrong options may be chosen. If the SSRE software detects an error during the export process, it will automatically stop execution and exit. The program will also automatically monitor the state of the export regardless of export speed and stop execution if an error is detected.
There are also a few additional options present only on specific sleep software. When exporting Noxturnal recordings, there is an option to automatically try to continue the export process if Noxturnal crashes which may occur when exporting very large datasets. If this option is selected, the SSRE software will attempt to automatically resume the export process when it detects Noxturnal crashing. The user can also select whether to attempt to force closing the old Noxturnal instance when recovering from a crash. In most cases, this option can be left unchecked and the extra Noxturnal instances can be simply closed manually after the export is complete. However, if the computer running the exports has a low amount of RAM, it may be better to enable this option. When exporting Profusion PSG recordings, there is an additional option to perform an automatic desaturation scoring analysis before exporting the events. If this option is used, the user should be aware that this automatic scoring will not replace the old scoring so this option should only be used on data that has no previous desaturation scoring saved as this would result in duplicate desaturation events. There are also additional options for all sleep scoring software to display additional information during the export and generate an info file for each subject during the export. These options are intended for troubleshooting and can be ignored normally.
When all desired settings are chosen, the export can be started by clicking “Start”. The final pop-up will appear that reminds the user to not touch the keyboard or mouse during the export as this could break the export. However, the user can press “Esc” at any time to immediately abort the export and close the program. Clicking “OK” will open the corresponding export program and start exporting. Once the export is finished for all selected recordings, the SSRE software will inform the user and return to the export selection window.
After the exports have been completed, the anonymization tool can be used to remove all patient information from the exports. However, this anonymization step is completely optional and usually only required if the data will be shared. In addition, some exports may not require the anonymization process at all as the exports do not include patient information in the first place. It is also possible to also anonymize data that is not exported using the SSRE software as long as the exported data is in the same formats that the SSRE would also output. If the user wishes to perform the data anonymization, the anonymization tool can be launched from the main selection window.
Once the tool launches, it opens the anonymization selection window (Fig. 5). From here, the user selects which program the exports were associated with, and which folder the exports are located in. If the default directories are used, the program will automatically use the same export folder as defined in the options. It should be noted that the program will perform the anonymization directly on the exports, and will not make another copy of them. Therefore, if the user wishes to back up the exports that still include the patient information, it should be done manually before the anonymization process. In addition, all export files in the folders will be anonymized. The program will automatically detect which files are present and sequentially anonymize them. Therefore, depending on which files were exported and the performance of the computer, the anonymization process can take from only a few ms/subject to as long as several minutes per subject.
Once the correct folder is selected, the user can press “Start” to begin the anonymization process, which will also open a progress window (Fig. 6). This window can be used to monitor the progress of the anonymization and the estimated time remaining. This window can also be used to prematurely stop the anonymization by pressing the “Stop” button. To ensure that the data is not corrupted by the stop-action, the anonymization of the current file will be finished and thus, the anonymization will not stop instantly. Therefore, the user should not attempt to force close the anonymization tool as this can lead to data corruption if the current write process is not allowed to complete. Once the “Close” button has been enabled, the window can be safely closed. When the anonymization is complete, the program will inform the user and at that point, the “Close” button will be again enabled and the program can be closed. It should be noted that the anonymization tool will not modify the names of the subject folders (12801–12810 in these examples). Therefore, the exports can still be linked to the original data and can only be considered pseudonymized.