Dataset Information

Extracting duration information in a picture category decoding task using hidden Markov Models.

ABSTRACT: Adapting classifiers for the purpose of brain signal decoding is a major challenge in brain-computer-interface (BCI) research. In a previous study we showed in principle that hidden Markov models (HMM) are a suitable alternative to the well-studied static classifiers. However, since we investigated a rather straightforward task, advantages from modeling of the signal could not be assessed.Here, we investigate a more complex data set in order to find out to what extent HMMs, as a dynamic classifier, can provide useful additional information. We show for a visual decoding problem that besides category information, HMMs can simultaneously decode picture duration without an additional training required. This decoding is based on a strong correlation that we found between picture duration and the behavior of the Viterbi paths.Decoding accuracies of up to 80% could be obtained for category and duration decoding with a single classifier trained on category information only.The extraction of multiple types of information using a single classifier enables the processing of more complex problems, while preserving good training results even on small databases. Therefore, it provides a convenient framework for online real-life BCI utilizations.

SUBMITTER: Pfeiffer T

PROVIDER: S-EPMC4871607 | biostudies-literature | 2016 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Extracting duration information in a picture category decoding task using hidden Markov Models.

Pfeiffer Tim T Heinze Nicolai N Frysch Robert R Deouell Leon Y LY Schoenfeld Mircea A MA Knight Robert T RT Rose Georg G

Journal of neural engineering 20160209 2

<h4>Objective</h4>Adapting classifiers for the purpose of brain signal decoding is a major challenge in brain-computer-interface (BCI) research. In a previous study we showed in principle that hidden Markov models (HMM) are a suitable alternative to the well-studied static classifiers. However, since we investigated a rather straightforward task, advantages from modeling of the signal could not be assessed.<h4>Approach</h4>Here, we investigate a more complex data set in order to find out to what ...[more]

PMID: 26859831

Similar Datasets

Project description:BackgroundRiboswitches are a type of noncoding RNA that regulate gene expression by switching from one structural conformation to another on ligand binding. The various classes of riboswitches discovered so far are differentiated by the ligand, which on binding induces a conformational switch. Every class of riboswitch is characterized by an aptamer domain, which provides the site for ligand binding, and an expression platform that undergoes conformational change on ligand binding. The sequence and structure of the aptamer domain is highly conserved in riboswitches belonging to the same class. We propose a method for fast and accurate identification of riboswitches using profile Hidden Markov Models (pHMM). Our method exploits the high degree of sequence conservation that characterizes the aptamer domain.ResultsOur method can detect riboswitches in genomic databases rapidly and accurately. Its sensitivity is comparable to the method based on the Covariance Model (CM). For six out of ten riboswitch classes, our method detects more than 99.5% of the candidates identified by the much slower CM method while being several hundred times faster. For three riboswitch classes, our method detects 97-99% of the candidates relative to the CM method. Our method works very well for those classes of riboswitches that are characterized by distinct and conserved sequence motifs.ConclusionRiboswitches play a crucial role in controlling the expression of several prokaryotic genes involved in metabolism and transport processes. As more and more new classes of riboswitches are being discovered, it is important to understand the patterns of their intra and inter genomic distribution. Understanding such patterns will enable us to better understand the evolutionary history of these genetic regulatory elements. However, a complete picture of the distribution pattern of riboswitches will emerge only after accurate identification of riboswitches across genomes. We believe that the riboswitch detection method developed in this paper will aid in that process. The significant advantage in terms of speed, of our pHMM-based approach over the method based on CM allows us to scan entire databases (rather than 5'UTRs only) in a relatively short period of time in order to accurately identify riboswitch candidates.

Dataset Information

Extracting duration information in a picture category decoding task using hidden Markov Models.

Publications

Extracting duration information in a picture category decoding task using hidden Markov Models.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets