Luna Scripts

Lunascope can execute Luna scripts to automate processing. These docks are hidden by default, but can be raised by either the View menu, or Ctrl/Cmd-8 (console dock) and Ctrl/Cmd-9 (output dock).

A simple script

To apply the HEADERS and STATS command, i.e. similar to running

luna s.lst 1 -o out.db -s ' HEADERS & STATS sig=C3,C4 '

enter the script in the console (which can also save/load scripts):

and then press Execute. This will send any output to the output dock. For example, this table is the same as the command line

destrat out.db +STATS -r CH

You can transpose the table for easier viewing, as needed.

Exporting tables

You can select the whole table (click top left), or individual rows, columns or cells and either Copy those data to the clipboard or save them as a TSV file, by right-clicking in the table. The exported data will always contain column headers. If a filter is currently applied, only filtered rows will be exported.

---

A more involved case

Here we have an example script to detect NREM spindles and slow oscillations, which takes optional arguments, and adds annotations (reflecting spindle/SO events):

If we load this script and just press Execute we'll get an error message:

This correctly notes that ${s} (the signals to use) is required but has not been specified. These need to be specified in the parameter dock (ignore the top examples in this case, but not how variables ${s}, ${r} (the reference channels) and ${frqs} (the spindle frequencies) have been specified:

This is now the equivalent to writing

luna s.lst 1 -o out.db s=C3 r=A1,A2 frqs=11,15 < spindle.txt

(assuming the Luna script is called spindle.txt).

Now it should run properly (i.e. assuming channels C3, A1 and A2 exist, and there are NREM annotations) and give output as follows: here it shows spindle metrics for fast and slow (F = 15 and 11 Hz) for two channels (here, s=C3,C4 obviously, rather than a single channel):

Inspecting the script, we see it will generate annotations (SP11 and SP15 as well as SO) marking where detected spindles occur. After running the script, these will be attached to the in-memory instance, and can be viewed (or used in any subsequent Luna command, until Refresh is hit or a new EDF loaded):

Selecting the two EEG channels -- and applying sigma-filtering on one of them -- as well as the relevant annotations, you can explore the detected NREM transient events in this dataset:

Batch processing

Although Lunascope is centrally designed around viewing a single recording, it can sometimes be convenient to apply a script to multiple recordings grouped as a sample-list.

Importantly, note that for non-trivial batch processing applications it is strongly suggested that you use command-line Luna, specifically designed for batch processing, not the Lunascope interactive viewer, for a more flexible, powerful, reproducible and robust pipeline.

A batch job in Lunascope applies the current Luna script to all samples in the current sample list, iteratively. For example, consider we have this script, to enumate the number of NREM and REM obstructive apnea events, respectively:

TAG STG/NREM
MASK ifnot=N1,N2,N3 
ANNOTS annot=Obstructive_Apnea

TAG STG/REM
MASK ifnot=R 
ANNOTS annot=Obstructive_Apnea

Instead of clicking the Execute button (that applies the script to the current state of the attached EDF), use the top menu Project / Evaluate (project):

This loads each EDF sequentially, runs the job, and saves any output. Only after all jobs are finished, the console will update with the output:

When finished, the viewer will have the final record in the sample list loaded.

The outputs are collated and then displayed in a single, collated form. An ID column (based on the sample list ID) is added to each table to indicate to which individual those data belong:

Note that Lunascope does not parallelize jobs and will freeze the GUI while jobs run. As such, it can be difficult to see the progress of long-running jobs. Also, it will not automatically save all outputs. As noted, for these and other reasons, Lunascope is not the right tool for non-trivial batch jobs.