Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2010 Feb;11(2):115-26.
doi: 10.1631/jzus.B0910427.

A biologically inspired model for pattern recognition

Eduardo Gonzalez  1 , Hans Liljenström, Yusely Ruiz, Guang Li
Affiliations

A biologically inspired model for pattern recognition

Eduardo Gonzalez et al. J Zhejiang Univ Sci B. 2010 Feb.

Abstract

In this paper, a novel bionic model and its performance in pattern recognition are presented and discussed. The model is constructed from a bulb model and a three-layered cortical model, mimicking the main features of the olfactory system. The olfactory bulb and cortex models are connected by feedforward and feedback fibers with distributed delays. The Breast Cancer Wisconsin dataset consisting of data from 683 patients divided into benign and malignant classes is used to demonstrate the capacity of the model to learn and recognize patterns, even when these are deformed versions of the originally learned patterns. The performance of the novel model was compared with three artificial neural networks (ANNs), a back-propagation network, a support vector machine classifier, and a radial basis function classifier. All the ANNs and the olfactory bionic model were tested in a benchmark study of a standard dataset. Experimental results show that the bionic olfactory system model can learn and classify patterns based on a small training set and a few learning trials to reflect biological intelligence to some extent.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
The structure of our bionic model. For the bulb model during the active stage the sensory input (a) is modulated by excitatory mitral cells (M) and inhibitory granule cells (G). Olfactory information (b) passes from the olfactory bulb directly into top (Layer I) and middle (Layer II) layers on the cortex model via the lateral olfactory track (LOT). Feedback information (c) is sent up from middle cortical layer to bulbar granule nodes via medial olfactory track (MOT). The bionic model output (d) is taken from the cortical middle layer
Fig. 2
Fig. 2
Architecture of the cortical model. (a) General principle for connections between excitatory nodes (open circles) and inhibitory nodes (filled circles). Dashed lines indicate that connection strength decrease with distance. The model is taken from Liljenström (1991); (b) Overall structure of the model olfactory cortex with intralayer and interlayer connections. The arrows represent the input from the bulb model. The model is taken from Liljenström (1991); (c) The excitatory nodes in middle layer, which make connections with inhibitory nodes in upper or bottom layer; (d) Maximum range (dij ,max) of 3 mm for inhibitory-excitatory connections
Fig. 2
Fig. 2
Architecture of the cortical model. (a) General principle for connections between excitatory nodes (open circles) and inhibitory nodes (filled circles). Dashed lines indicate that connection strength decrease with distance. The model is taken from Liljenström (1991); (b) Overall structure of the model olfactory cortex with intralayer and interlayer connections. The arrows represent the input from the bulb model. The model is taken from Liljenström (1991); (c) The excitatory nodes in middle layer, which make connections with inhibitory nodes in upper or bottom layer; (d) Maximum range (dij ,max) of 3 mm for inhibitory-excitatory connections
Fig. 2
Fig. 2
Architecture of the cortical model. (a) General principle for connections between excitatory nodes (open circles) and inhibitory nodes (filled circles). Dashed lines indicate that connection strength decrease with distance. The model is taken from Liljenström (1991); (b) Overall structure of the model olfactory cortex with intralayer and interlayer connections. The arrows represent the input from the bulb model. The model is taken from Liljenström (1991); (c) The excitatory nodes in middle layer, which make connections with inhibitory nodes in upper or bottom layer; (d) Maximum range (dij ,max) of 3 mm for inhibitory-excitatory connections
Fig. 2
Fig. 2
Architecture of the cortical model. (a) General principle for connections between excitatory nodes (open circles) and inhibitory nodes (filled circles). Dashed lines indicate that connection strength decrease with distance. The model is taken from Liljenström (1991); (b) Overall structure of the model olfactory cortex with intralayer and interlayer connections. The arrows represent the input from the bulb model. The model is taken from Liljenström (1991); (c) The excitatory nodes in middle layer, which make connections with inhibitory nodes in upper or bottom layer; (d) Maximum range (dij ,max) of 3 mm for inhibitory-excitatory connections
Fig. 3
Fig. 3
Non-linear input-output functions for mitral (solid line) and granule (dashed line) nodes, based on Freeman and Skarda (1985)
Fig. 4
Fig. 4
Cortical areas (black) receiving input from olfactory bulb
Fig. 5
Fig. 5
Optimization of bionic model channels for BCW dataset
Fig. 6
Fig. 6
Optimization of spread parameter for the RBF
Fig. 7
Fig. 7
Optimization of the training instances per class for BCW dataset. (a) BPN; (b) SVM classifier; (c) RBF classifier; (d) OBM
Fig. 7
Fig. 7
Optimization of the training instances per class for BCW dataset. (a) BPN; (b) SVM classifier; (c) RBF classifier; (d) OBM
Fig. 7
Fig. 7
Optimization of the training instances per class for BCW dataset. (a) BPN; (b) SVM classifier; (c) RBF classifier; (d) OBM
Fig. 7
Fig. 7
Optimization of the training instances per class for BCW dataset. (a) BPN; (b) SVM classifier; (c) RBF classifier; (d) OBM
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Amaral DG, Insausti R, Cowan WM. The entorhinal cortex of the monkey: I. Cytoarchitectonic organization. The Journal of Comparative Neurology. 1987;264(3):326–355. doi: 10.1002/cne.902640305. - DOI - PubMed
    1. Aronsson P, Liljenström H. Effects of non-synaptic neuronal interaction in cortex on synchronization and learning. Biosystems. 2001;63(1-3):43–56. doi: 10.1016/S0303-2647(01)00146-0. - DOI - PubMed
    1. Brunzell H, Eriksson J. Feature reduction for classification of multidimensional data. Pattern Recognition. 2000;33(10):1741–1748. doi: 10.1016/S0031-3203(99)00142-9. - DOI
    1. Cortes C, Vapnik V. Support-vector networks. Machine Learning. 1995;20(3):273–297. doi: 10.1007/BF00994018. - DOI
    1. de Araujo IE, Rolls ET, Velazco MI, Margot C, Cayeux I. Cognitive modulation of olfactory processing. Neuron. 2005;46(4):671–679. doi: 10.1016/j.neuron.2005.04.021. - DOI - PubMed

Publication types