thenineteen / Semiology-Visualisation-Tool

Data driven 3D brain visualisation of semiology. Semiology to anatomy translator based on over 4600 patients from 309 peer-reviewed articles.
MIT License
9 stars 6 forks source link

GUI support for inverse localising spread value (ILV) option #191

Closed thenineteen closed 3 years ago

thenineteen commented 4 years ago

related to pull #190

Please can we have an option right under the Granular option in settings (tab indented) that states Use Inverse Localising Spread Values.

thenineteen commented 4 years ago

the tool tip should read

"This option, when enabled, reduces the localising-values in the Semio2Brain database by a factor inversely proportional to the number of brain regions the semiology of interest localised to.

In effect, this option provides an inverse variance weighting relative to the spread of localisation, favouring semiologies which are more (uni)-focal.

This option is only available when using the granular postcode hierarchy reversal option."

fepegar commented 3 years ago

Done, but I get an error:

Error retrieving semiology information from mega_analysis module. Details:

cannot convert the series to <class 'float'>

If you think this is a bug, please report this issue on the repository: https://github.com/thenineteen/Semiology-Visualisation-Tool/issues/new

Are there tests for that option?

fepegar commented 3 years ago

P.S.: please stop changing the indentation in semiology.py. Your editor is doing it when you save.

thenineteen commented 3 years ago

Thanks for implementing this in the GUI!

Can you clarify what gives you that error? At what point in the query: do all semiology queries guve that error or a specific case?

I saved without formatting on semiology.py but maybe I accidentally habitually pressed ctrl+s at some point, apologies if there are formatting changes.

There are indeed tests for this here https://github.com/thenineteen/Semiology-Visualisation-Tool/blob/master/tests/test_inverse_localising_values.py

fepegar commented 3 years ago

Sorry, I clicked on the new checkbox and queried Tonic - left.

fepegar commented 3 years ago

Easiest way to reproduce: python -c "import mega_analysis; np = mega_analysis.Semiology('Tonic', symptoms_side=mega_analysis.semiology.Laterality.LEFT, dominant_hemisphere=mega_analysis.semiology.Laterality.LEFT, inverse_localising_values=True).get_num_datapoints_dict().

But this is a different issue, so I'll close this.