Functional magnetic resonance imaging (fMRI) generates vast amounts of data. The handling, processing, and analysis of fMRI data would be inconceivable without computer-based methods. fMRI neuroinformatics is concerned with research, development, and operation of these methods. Reconstruction, rudimentary analysis, and visualization tools are implemented in software controlling modern MRI scanners. Research in advanced methods for analysis of subtle activation patterns, realistic physiological modeling, or integration of data from multiple subjects is the basis for a lively research field and has led to the development of a large number of tools.

http://ieeexplore.ieee.org/iel5/51/33759/01607675.pdf

fMRI neuroinformatics

The electronic linking of neuroscience information, including data embedded in the primary literature, would permit powerful queries and analyses driven by structured databases. This task would be facilitated by automated procedures which can identify biological concepts in journals. Here we apply an approach for automatically mapping formal identifiers of neuroanatomical regions to text found in journal abstracts, and apply it to a large body of abstracts from the Journal of Comparative Neurology (JCN). The analyses yield over one hundred thousand brain region mentions which we map to 8,225 brain region concepts in multiple organisms. Based on the analysis of a manually annotated corpus, we estimate mentions are mapped at 95% precision and 63% recall. Our results provide insights into the patterns of publication on brain regions and species of study in the Journal, but also point to important challenges in the standardization of neuroanatomical nomenclatures. We find that many terms in the formal terminologies never appear in a JCN abstract, while conversely, many terms authors use are not reflected in the terminologies. To improve the terminologies we deposited 136 unrecognized brain regions into the Neuroscience Lexicon (NeuroLex). The training data, terminologies, normalizations, evaluations and annotated journal abstracts are freely available at http://www.chibi.ubc.ca/WhiteText/.

/pdf/using-text-mining-to-link-journal-articles-to-1sf7k1vk7f.pdf

Using text mining to link journal articles to neuroanatomical databases

The wealth and diversity of neuroscience research are inherent characteristics of the discipline that can give rise to some complications. As the field continues to expand, we generate a great deal of data about all aspects, and from multiple perspectives, of the brain, its chemistry, biology, and how these affect behavior. The vast majority of research scientists cannot afford to spend their time combing the literature to find every article related to their research, nor do they wish to spend time adjusting their neuroanatomical vocabulary to communicate with other subdomains in the neurosciences. As such, there has been a recent increase in the amount of informatics research devoted to developing digital resources for neuroscience research. Neuroinformatics is concerned with the development of computational tools to further our understanding of the brain and to make sense of the vast amount of information that neuroscientists generate (French & Pavlidis, 2007). Many of these tools are related to the use of textual data. Here, we review some of the recent developments for better using the vast amount of textual information generated in neuroscience research and publication and suggest several use cases that will demonstrate how bench neuroscientists can take advantage of the resources that are available.

Text-mining and neuroscience.

The emphasis of multilevel modeling techniques in the Neurosciences has led to an increased need for large-scale databases containing neuroscientific data. Despite this, such databases are not being populated at a rate commensurate with their demand amongst Computational Neuroscientists. The reasons for this are common to scientific database curation in general–limitation of resources. Much of Neuroscience’s long tradition of research has been documented in computationally inaccessible formats, such as the pdf, making large-scale data extraction laborious and expensive. Here we present three sets of studies designed to construct automated tools for alleviating three bottlenecks in the workflow of a community-curated knowledge base of neuroscience-related information. Virk, the first of these tools, is designed specifically with under-developed knowledge bases in mind, using active learning to allow them to quickly bootstrap their development. Flokka, our second tool, is designed for prioritizing a set of potentially-relevant manuscripts, so that they can be examined in order of their likely relevance. Finna, our final tool, is designed to rank-order the composite paragraphs of a likely relevant document, in terms of the probability that the paragraph contains information that is of interest to the database. Each of our systems attained a level of performance indicating its potential usefulness in the real world. In addition, we present a data set consisting of 962 expertcurated neuroscience documents–to our knowledge, the first data set of its kind. Each document was annotated at the document level, in terms of their relevance for a neuron-related knowledge base, and at the sentence level, in terms of whether a particular sentence communicates information that would lead to the document being included in the knowledge base.

Text-mining tools for optimizing community database curation workflows in neuroscience

Bioinformatics for neuroanatomical connectivity

Generating informational thesauri that classify, cross-reference, and retrieve diverse and highly detailed neuroscientific information requires identifying related neuroanatomical terms and acronyms within and between species (Gorin et al., 2001) Manual construction of such informational thesauri is laborious, and we describe implementing and evaluating a neuroanatomical term and acronym reconciliation (NTAR) system to assist domain experts with this task. NTAR is composed of two modules. The neuroanatomical term extraction (NTE) module employs a hidden Markov model (HMM) in conjunction with lexical rules to extract neuroanatomical terms (NT) and acronyms (NA) from textual material. The output of the NTE is formatted into collections of term- or acronym-indexed documents composed of sentences and word phrases extracted from textual material. The second information retrieval (IR) module utilizes a vector space model (VSM) and includes a novel, automated relevance feedback algorithm. The IR module retrieves statistically related neuroanatomical terms and acronyms in response to queried neuroanatomical terms and acronyms. Neuroanatomical terms and acronyms retrieval obtained from term-based inquiries were compared with (1) term retrieval obtained by including automated relevance feedback and with (2) term retrieval using “document-to-document” comparisons (context-based VSM). The retrieval of synonymous and similar primate and macaque thalamic terms and acronyms in response to a query list of human thalamic terminology by these three IR approaches was compared against a previously published, manually constructed concordance table of homologous cross-species terms and acronyms. Term-based VSM with automated relevance feedback retrieved 70% and 80% of these primate and macaque terms and acronyms, respectively, listed in the concordance table. Automated feedback algorithm correctly identified 87% of the macaque terms and acronyms that were independently selected by a domain expert as being appropriate for manual relevance feedback. Context-based VSM correctly retrieved 97% and 98% of the primate and macaque terms and acronyms listed in the term homology table. These results indicate that the NTAR system could assist neuroscientists with thesauri creation for closely related, highly detailed neuroanatomical domains.

Shang-Heng Wei

Papers

Comparison of vector space model methodologies to reconcile cross-species neuroanatomical concepts.