Discovering relations between mind, brain, and mental disorders using topic mapping

From Brede Wiki
(Redirected from PMID/23071428)
Jump to: navigation, search
Paper (help)
Discovering relations between mind, brain, and mental disorders using topic mapping
Authors: Russell A. Poldrack, Jeanette A. Mumford, Tom Schonberg, Donald Kalar, Bishal Barman, Tal Yarkoni
Citation: PLoS Computational Biology 8 (10): e1002707. 2012
Database(s): Google Scholar cites PubMed (PMID/23071428)
DOI: 10.1371/journal.pcbi.1002707.
PMCID:3469446
Link(s): http://talyarkoni.com/papers/Poldrack_et_al_in_press_PLoS_CompBio.pdf
Search
Web: Bing Google Yahoo!Google PDF
Article: BASE Google Scholar PubMed
Restricted: DTU Digital Library
Other: NIF
Services
Format: BibTeX Template from PMID
Extract: Talairach coordinates from linked PDF: CSV-formated wiki-formated

Discovering relations between mind, brain, and mental disorders using topic mapping

Software for the study is available at

https://github.com/poldrack/LatentStructure

There is supplementary material:

This paper is related to an earlier abstract Topic mapping: a literature-wide analysis of mind-brain relationships.

Contents

[edit] Abstract (CC-BY)

Neuroimaging research has largely focused on the identification of associations between brain activation and specific mental functions. Here we show that data mining techniques applied to a large database of neuroimaging results can be used to identify the conceptual structure of mental functions and their mapping to brain systems. This analysis confirms many current ideas regarding the neural organization of cognition, but also provides some new insights into the roles of particular brain systems in mental function. We further show that the same methods can be used to identify the relations between mental disorders. Finally, we show that these two approaches can be combined to empirically identify novel relations between mental disorders and mental functions via their common involvement of particular brain networks. This approach has the potential to discover novel endophenotypes for neuropsychiatric disorders and to better characterize the structure of these disorders and the relations between them.

[edit] Method

  • Data was obtained from Neurosynth. The version with 5,809 articles was used.
  • Cognitive Atlas was used for a positive list of words. 605 terms was used for mental functions and 55 for mental disorders
  • Topic modeling the latent Diriclet allocation with an implementation in MALLET.
  • The number of topics in the topic model is determined from the empirical likelihood
  • Construction of images by "placing a sphere (10 mm radius) at each activation location, at 3 mm resolution using the MNI305 template."
  • Activation converted to MNI305 with the Lancaster transform.[2]
  • voxelwise chi-squared p-value maps between topic maps and constructed images.

[edit] Python

[edit] Get terms from topics

from urllib import urlretrieve
from subprocess import call
from codecs import open
from re import compile, sub, UNICODE
 
urlretrieve('http://s3-eu-west-1.amazonaws.com/files.figshare.com/298932/Table_S1.pdf', 'Table_S1.pdf')
call(['pdftotext', 'Table_S1.pdf'])
table = {ord('_'): ord(' ')}
terms = [term.strip().translate(table) for line in open('Table_S1.txt', encoding='utf-8') 
                                       for term in line.split(',') if ',' in line]
 
# Kerning issue in PDF translation
kerning_chars = set([char for term in terms for char in term if not 'A' <= char <= 'z'])
translation_table = {u'\ufb01': 'fi', u'\ufb02': 'fl'}
pattern = compile("(" + r"|".join(translation_table) + ")", flags=UNICODE)
translator = lambda match: translation_table[match.string[match.start():match.end()]]
terms = set([pattern.sub(translator, term) for term in terms])

[edit] Compare with Brede

https://github.com/fnielsen/brede/

from brede.data.words import CognitiveWords
 
terms - CognitiveWords()

[edit] Related papers

  1. Mining the posterior cingulate: segregation between memory and pain components
  2. Mining for associations between text and brain activation in a functional neuroimaging database

[edit] See also

[edit] References

  1. A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis
  2. Bias between MNI and Talairach coordinates analyzed using the ICBM-152 brain template
Personal tools