Improve functionality of ElectrodeCategoryReader

[ariel215]

The electrode category file is highly irregular, but we do need to be able to consistently read it in a reasonable form

[mivade]

What's wrong with the current reader for it? It handles a lot of the irregularities.

cmlreaders/cmlreaders/readers/readers.py

Lines 246 to 252 in 2e96eda

	class ElectrodeCategoriesReader(BaseCMLReader):
	"""Reads electrode_categories.txt and handles the many inconsistencies in
	those files.
	Returns a ``dict``.
	"""

[esolomon]

@mivade Having looked at my personal code that handles electrode categories, I think providing you with some pseudocode would honestly be more useful:

For each line of electrode_categories.txt to be "decoded":
if 'seizure' in txt.lower() or 'soz' in txt.lower() or 'onset' in txt.lower():
Classify as "Seizure Onset Zone"

if 'ictal' in txt.lower():
Classify as 'Interictal Spiking'

if 'bad' in txt.lower() or 'broken' in txt.lower():
Classify as "Bad Electrode"

if 'lesion' in txt.lower():
Classify as "Brain Lesion"

In other words, search for ('seizure', 'soz') to find Seizure Onset Zone labels, search for ('ictal') to find Interictal Spiking electrodes, and search for ('broken', 'bad') to find Bad Electrodes.