Detecting bladder fullness through the ensemble activity patterns of the spinal cord unit population in a somatovisceral convergence environment

Jae Hong Park, Chang Eop Kim, Jaewoo Shin, Changkyun Im, Chin Su Koh, In Seok Seo, Sangjeong Kim, Hyung Cheul Shin

Research output: Contribution to journalArticleResearchpeer-review

5 Citations (Scopus)

Abstract

Objective. Chronic monitoring of the state of the bladder can be used to notify patients with urinary dysfunction when the bladder should be voided. Given that many spinal neurons respond both to somatic and visceral inputs, it is necessary to extract bladder information selectively from the spinal cord. Here, we hypothesize that sensory information with distinct modalities should be represented by the distinct ensemble activity patterns within the neuronal population and, therefore, analyzing the activity patterns of the neuronal population could distinguish bladder fullness from somatic stimuli. Approach. We simultaneously recorded 26-27 single unit activities in response to bladder distension or tactile stimuli in the dorsal spinal cord of each Sprague-Dawley rat. In order to discriminate between bladder fullness and tactile stimulus inputs, we analyzed the ensemble activity patterns of the entire neuronal population. A support vector machine (SVM) was employed as a classifier, and discrimination performance was measured by k-fold cross-validation tests. Main results. Most of the units responding to bladder fullness also responded to the tactile stimuli (88.9-100%). The SVM classifier precisely distinguished the bladder fullness from the somatic input (100%), indicating that the ensemble activity patterns of the unit population in the spinal cord are distinct enough to identify the current input modality. Moreover, our ensemble activity pattern-based classifier showed high robustness against random losses of signals. Significance. This study is the first to demonstrate that the two main issues of electroneurographic monitoring of bladder fullness, low signals and selectiveness, can be solved by an ensemble activity pattern-based approach, improving the feasibility of chronic monitoring of bladder fullness by neural recording.

Original languageEnglish
Article number056009
JournalJournal of Neural Engineering
Volume10
Issue number5
DOIs
StatePublished - 1 Oct 2013

Fingerprint

Spinal Cord
Urinary Bladder
Classifiers
Support vector machines
Monitoring
Population
Touch
Neurons
Rats
Sprague Dawley Rats

Cite this

Park, Jae Hong ; Kim, Chang Eop ; Shin, Jaewoo ; Im, Changkyun ; Koh, Chin Su ; Seo, In Seok ; Kim, Sangjeong ; Shin, Hyung Cheul. / Detecting bladder fullness through the ensemble activity patterns of the spinal cord unit population in a somatovisceral convergence environment. In: Journal of Neural Engineering. 2013 ; Vol. 10, No. 5.
@article{7ece36162e9e43f4a46f7910be529f08,
title = "Detecting bladder fullness through the ensemble activity patterns of the spinal cord unit population in a somatovisceral convergence environment",
abstract = "Objective. Chronic monitoring of the state of the bladder can be used to notify patients with urinary dysfunction when the bladder should be voided. Given that many spinal neurons respond both to somatic and visceral inputs, it is necessary to extract bladder information selectively from the spinal cord. Here, we hypothesize that sensory information with distinct modalities should be represented by the distinct ensemble activity patterns within the neuronal population and, therefore, analyzing the activity patterns of the neuronal population could distinguish bladder fullness from somatic stimuli. Approach. We simultaneously recorded 26-27 single unit activities in response to bladder distension or tactile stimuli in the dorsal spinal cord of each Sprague-Dawley rat. In order to discriminate between bladder fullness and tactile stimulus inputs, we analyzed the ensemble activity patterns of the entire neuronal population. A support vector machine (SVM) was employed as a classifier, and discrimination performance was measured by k-fold cross-validation tests. Main results. Most of the units responding to bladder fullness also responded to the tactile stimuli (88.9-100{\%}). The SVM classifier precisely distinguished the bladder fullness from the somatic input (100{\%}), indicating that the ensemble activity patterns of the unit population in the spinal cord are distinct enough to identify the current input modality. Moreover, our ensemble activity pattern-based classifier showed high robustness against random losses of signals. Significance. This study is the first to demonstrate that the two main issues of electroneurographic monitoring of bladder fullness, low signals and selectiveness, can be solved by an ensemble activity pattern-based approach, improving the feasibility of chronic monitoring of bladder fullness by neural recording.",
author = "Park, {Jae Hong} and Kim, {Chang Eop} and Jaewoo Shin and Changkyun Im and Koh, {Chin Su} and Seo, {In Seok} and Sangjeong Kim and Shin, {Hyung Cheul}",
year = "2013",
month = "10",
day = "1",
doi = "10.1088/1741-2560/10/5/056009",
language = "English",
volume = "10",
journal = "Journal of Neural Engineering",
issn = "1741-2560",
publisher = "IOP Publishing Ltd.",
number = "5",

}

Detecting bladder fullness through the ensemble activity patterns of the spinal cord unit population in a somatovisceral convergence environment. / Park, Jae Hong; Kim, Chang Eop; Shin, Jaewoo; Im, Changkyun; Koh, Chin Su; Seo, In Seok; Kim, Sangjeong; Shin, Hyung Cheul.

In: Journal of Neural Engineering, Vol. 10, No. 5, 056009, 01.10.2013.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Detecting bladder fullness through the ensemble activity patterns of the spinal cord unit population in a somatovisceral convergence environment

AU - Park, Jae Hong

AU - Kim, Chang Eop

AU - Shin, Jaewoo

AU - Im, Changkyun

AU - Koh, Chin Su

AU - Seo, In Seok

AU - Kim, Sangjeong

AU - Shin, Hyung Cheul

PY - 2013/10/1

Y1 - 2013/10/1

N2 - Objective. Chronic monitoring of the state of the bladder can be used to notify patients with urinary dysfunction when the bladder should be voided. Given that many spinal neurons respond both to somatic and visceral inputs, it is necessary to extract bladder information selectively from the spinal cord. Here, we hypothesize that sensory information with distinct modalities should be represented by the distinct ensemble activity patterns within the neuronal population and, therefore, analyzing the activity patterns of the neuronal population could distinguish bladder fullness from somatic stimuli. Approach. We simultaneously recorded 26-27 single unit activities in response to bladder distension or tactile stimuli in the dorsal spinal cord of each Sprague-Dawley rat. In order to discriminate between bladder fullness and tactile stimulus inputs, we analyzed the ensemble activity patterns of the entire neuronal population. A support vector machine (SVM) was employed as a classifier, and discrimination performance was measured by k-fold cross-validation tests. Main results. Most of the units responding to bladder fullness also responded to the tactile stimuli (88.9-100%). The SVM classifier precisely distinguished the bladder fullness from the somatic input (100%), indicating that the ensemble activity patterns of the unit population in the spinal cord are distinct enough to identify the current input modality. Moreover, our ensemble activity pattern-based classifier showed high robustness against random losses of signals. Significance. This study is the first to demonstrate that the two main issues of electroneurographic monitoring of bladder fullness, low signals and selectiveness, can be solved by an ensemble activity pattern-based approach, improving the feasibility of chronic monitoring of bladder fullness by neural recording.

AB - Objective. Chronic monitoring of the state of the bladder can be used to notify patients with urinary dysfunction when the bladder should be voided. Given that many spinal neurons respond both to somatic and visceral inputs, it is necessary to extract bladder information selectively from the spinal cord. Here, we hypothesize that sensory information with distinct modalities should be represented by the distinct ensemble activity patterns within the neuronal population and, therefore, analyzing the activity patterns of the neuronal population could distinguish bladder fullness from somatic stimuli. Approach. We simultaneously recorded 26-27 single unit activities in response to bladder distension or tactile stimuli in the dorsal spinal cord of each Sprague-Dawley rat. In order to discriminate between bladder fullness and tactile stimulus inputs, we analyzed the ensemble activity patterns of the entire neuronal population. A support vector machine (SVM) was employed as a classifier, and discrimination performance was measured by k-fold cross-validation tests. Main results. Most of the units responding to bladder fullness also responded to the tactile stimuli (88.9-100%). The SVM classifier precisely distinguished the bladder fullness from the somatic input (100%), indicating that the ensemble activity patterns of the unit population in the spinal cord are distinct enough to identify the current input modality. Moreover, our ensemble activity pattern-based classifier showed high robustness against random losses of signals. Significance. This study is the first to demonstrate that the two main issues of electroneurographic monitoring of bladder fullness, low signals and selectiveness, can be solved by an ensemble activity pattern-based approach, improving the feasibility of chronic monitoring of bladder fullness by neural recording.

UR - http://www.scopus.com/inward/record.url?scp=84885464247&partnerID=8YFLogxK

U2 - 10.1088/1741-2560/10/5/056009

DO - 10.1088/1741-2560/10/5/056009

M3 - Article

VL - 10

JO - Journal of Neural Engineering

JF - Journal of Neural Engineering

SN - 1741-2560

IS - 5

M1 - 056009

ER -