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 Apr;11(4):298-306.
doi: 10.1631/jzus.B0900284.

Neural decoding based on probabilistic neural network

Yi Yu  1 , Shao-min Zhang, Huai-jian Zhang, Xiao-chun Liu, Qiao-sheng Zhang, Xiao-xiang Zheng, Jian-hua Dai
Affiliations

Neural decoding based on probabilistic neural network

Yi Yu et al. J Zhejiang Univ Sci B. 2010 Apr.

Abstract

Brain-machine interface (BMI) has been developed due to its possibility to cure severe body paralysis. This technology has been used to realize the direct control of prosthetic devices, such as robot arms, computer cursors, and paralyzed muscles. A variety of neural decoding algorithms have been designed to explore relationships between neural activities and movements of the limbs. In this paper, two novel neural decoding methods based on probabilistic neural network (PNN) in rats were introduced, the PNN decoder and the modified PNN (MPNN) decoder. In the experiment, rats were trained to obtain water by pressing a lever over a pressure threshold. Microelectrode array was implanted in the motor cortex to record neural activity, and pressure was recorded by a pressure sensor synchronously. After training, the pressure values were estimated from the neural signals by PNN and MPNN decoders. Their performances were evaluated by a correlation coefficient (CC) and a mean square error (MSE). The results show that the MPNN decoder, with a CC of 0.8657 and an MSE of 0.2563, outperformed the traditionally-used Wiener filter (WF) and Kalman filter (KF) decoders. It was also observed that the discretization level did not affect the MPNN performance, indicating that the MPNN decoder can handle different tasks in BMI system, including the detection of movement states and estimation of continuous kinematic parameters.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
Probabilistic neural network structure Xj: input vector; Mi: number of training vector in class Ci; Si: summation of probability density in class Ci; D: decision of the model
Fig. 2
Fig. 2
Simultaneously recorded neural signal and pressure signal (a) Two-channel raw neuronal signals in rat S9-03’s motor cortex during 5 s; (b) Sorted spikes in two channels respectively; (c) Raster of neural ensemble firing. Each vertical short line in the figure denoted a spike firing; (d) Binned neuronal spike firing. Each row represented a neuron, and each column represented a bin period, where bin size was 100 ms. The color denoted the firing frequency; (e) Pressure value of the lever from pressure sensor. An abrupt change of the pressure value indicates a pressing event of rat. When the rat did not press the lever, the pressure value rested in the base line; (f) Discretized pressure when the bin size was 100 ms and the discretization level was 100
Fig. 2
Fig. 2
Simultaneously recorded neural signal and pressure signal (a) Two-channel raw neuronal signals in rat S9-03’s motor cortex during 5 s; (b) Sorted spikes in two channels respectively; (c) Raster of neural ensemble firing. Each vertical short line in the figure denoted a spike firing; (d) Binned neuronal spike firing. Each row represented a neuron, and each column represented a bin period, where bin size was 100 ms. The color denoted the firing frequency; (e) Pressure value of the lever from pressure sensor. An abrupt change of the pressure value indicates a pressing event of rat. When the rat did not press the lever, the pressure value rested in the base line; (f) Discretized pressure when the bin size was 100 ms and the discretization level was 100
Fig. 2
Fig. 2
Simultaneously recorded neural signal and pressure signal (a) Two-channel raw neuronal signals in rat S9-03’s motor cortex during 5 s; (b) Sorted spikes in two channels respectively; (c) Raster of neural ensemble firing. Each vertical short line in the figure denoted a spike firing; (d) Binned neuronal spike firing. Each row represented a neuron, and each column represented a bin period, where bin size was 100 ms. The color denoted the firing frequency; (e) Pressure value of the lever from pressure sensor. An abrupt change of the pressure value indicates a pressing event of rat. When the rat did not press the lever, the pressure value rested in the base line; (f) Discretized pressure when the bin size was 100 ms and the discretization level was 100
Fig. 2
Fig. 2
Simultaneously recorded neural signal and pressure signal (a) Two-channel raw neuronal signals in rat S9-03’s motor cortex during 5 s; (b) Sorted spikes in two channels respectively; (c) Raster of neural ensemble firing. Each vertical short line in the figure denoted a spike firing; (d) Binned neuronal spike firing. Each row represented a neuron, and each column represented a bin period, where bin size was 100 ms. The color denoted the firing frequency; (e) Pressure value of the lever from pressure sensor. An abrupt change of the pressure value indicates a pressing event of rat. When the rat did not press the lever, the pressure value rested in the base line; (f) Discretized pressure when the bin size was 100 ms and the discretization level was 100
Fig. 2
Fig. 2
Simultaneously recorded neural signal and pressure signal (a) Two-channel raw neuronal signals in rat S9-03’s motor cortex during 5 s; (b) Sorted spikes in two channels respectively; (c) Raster of neural ensemble firing. Each vertical short line in the figure denoted a spike firing; (d) Binned neuronal spike firing. Each row represented a neuron, and each column represented a bin period, where bin size was 100 ms. The color denoted the firing frequency; (e) Pressure value of the lever from pressure sensor. An abrupt change of the pressure value indicates a pressing event of rat. When the rat did not press the lever, the pressure value rested in the base line; (f) Discretized pressure when the bin size was 100 ms and the discretization level was 100
Fig. 2
Fig. 2
Simultaneously recorded neural signal and pressure signal (a) Two-channel raw neuronal signals in rat S9-03’s motor cortex during 5 s; (b) Sorted spikes in two channels respectively; (c) Raster of neural ensemble firing. Each vertical short line in the figure denoted a spike firing; (d) Binned neuronal spike firing. Each row represented a neuron, and each column represented a bin period, where bin size was 100 ms. The color denoted the firing frequency; (e) Pressure value of the lever from pressure sensor. An abrupt change of the pressure value indicates a pressing event of rat. When the rat did not press the lever, the pressure value rested in the base line; (f) Discretized pressure when the bin size was 100 ms and the discretization level was 100
Fig. 3
Fig. 3
Examples of pressure decoding results (rat S9-03) (a) MPNN; (b) PNN; (c) KF; (d) WF
Fig. 4
Fig. 4
Effect of discretization level on MPNN decoder (a) Performances of MPNN decoder in a range of discretization level of three rats; (b) Waveforms of MPNN outputs in a range of discretization levels of three rats
Fig. 4
Fig. 4
Effect of discretization level on MPNN decoder (a) Performances of MPNN decoder in a range of discretization level of three rats; (b) Waveforms of MPNN outputs in a range of discretization levels of three rats
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Brockwell AE, Rojas AL, Kass RE. Recursive bayesian decoding of motor cortical signals by particle filtering. J Neurophysiol. 2004;91(4):1899–1907. doi: 10.1152/jn.00438.2003. - DOI - PubMed
    1. Brown EN, Frank LM, Tang D, Quirk MC, Wilson MA. A statistical paradigm for neural spike train decoding applied to position prediction from ensemble firing patterns of rat hippocampal place cells. J Neurosci. 1998;18(18):7411–7425. - PMC - PubMed
    1. Carmena JM, Lebedev MA, Crist RE, O(Doherty JE, Santucci DM. Learning to control a brain-machine interface for reaching and grasping by primates. PLoS Biol. 2003;1(2):E42. doi: 10.1371/journal.pbio.0000042. - DOI - PMC - PubMed
    1. Eden UT, Frank LM, Barbieri R, Solo V, Brown EN. Dynamic analysis of neural encoding by point process adaptive filtering. Neural Comput. 2004;16(5):971–998. doi: 10.1162/089976604773135069. - DOI - PubMed
    1. Feng ZY, Chen WD, Ye XS. A remote control training system for rat navigation in complicated environment. J Zhejiang Univ-Sci A. 2007;8(2):323–330. doi: 10.1631/jzus.2007.A0323. - DOI

Publication types