Time series prediction of lung cancer patients' breathing pattern based on nonlinear dynamics

R. P. Tolakanahalli, D. K. Tewatia, Wolfgang A. Tome

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This study focuses on predicting breathing pattern, which is crucial to deal with system latency in the treatments of moving lung tumors. Predicting respiratory motion in real-time is challenging, due to the inherent chaotic nature of breathing patterns, i.e. sensitive dependence on initial conditions. In this work, nonlinear prediction methods are used to predict the short-term evolution of the respiratory system for 62 patients, whose breathing time series was acquired using respiratory position management (RPM) system. Single step and N-point multi step prediction are performed for sampling rates of 5Hz and 10Hz. We compare the employed non-linear prediction methods with respect to prediction accuracy to Adaptive Infinite Impulse Response (IIR) prediction filters. A Local Average Model (LAM) and local linear models (LLMs) combined with a set of linear regularization techniques to solve ill-posed regression problems are implemented. For all sampling frequencies both single step and N-point multi step prediction results obtained using LAM and LLM with regularization methods perform better than IIR prediction filters for the selected sample patients. Moreover, since the simple LAM model performs as well as the more complicated LLM models in our patient sample, its use for non-linear prediction is recommended.

Original languageEnglish (US)
Pages (from-to)257-265
Number of pages9
JournalPhysica Medica
Volume31
Issue number3
DOIs
StatePublished - May 1 2015

Fingerprint

Nonlinear Dynamics
breathing
lungs
Linear Models
Lung Neoplasms
Respiration
cancer
predictions
Respiratory System
impulses
Lung
sampling
respiratory system
filters
management systems
Neoplasms
regression analysis
tumors
Therapeutics

Keywords

  • Breathing pattern
  • Embedding dimension
  • Infinite response prediction filters
  • Lung cancer
  • Non-linear dynamics
  • Non-linear prediction
  • Prediction
  • Time delay

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging
  • Physics and Astronomy(all)

Cite this

Time series prediction of lung cancer patients' breathing pattern based on nonlinear dynamics. / Tolakanahalli, R. P.; Tewatia, D. K.; Tome, Wolfgang A.

In: Physica Medica, Vol. 31, No. 3, 01.05.2015, p. 257-265.

Research output: Contribution to journalArticle

@article{441a5fd06bbc40fab0a32d66efb8f29c,
title = "Time series prediction of lung cancer patients' breathing pattern based on nonlinear dynamics",
abstract = "This study focuses on predicting breathing pattern, which is crucial to deal with system latency in the treatments of moving lung tumors. Predicting respiratory motion in real-time is challenging, due to the inherent chaotic nature of breathing patterns, i.e. sensitive dependence on initial conditions. In this work, nonlinear prediction methods are used to predict the short-term evolution of the respiratory system for 62 patients, whose breathing time series was acquired using respiratory position management (RPM) system. Single step and N-point multi step prediction are performed for sampling rates of 5Hz and 10Hz. We compare the employed non-linear prediction methods with respect to prediction accuracy to Adaptive Infinite Impulse Response (IIR) prediction filters. A Local Average Model (LAM) and local linear models (LLMs) combined with a set of linear regularization techniques to solve ill-posed regression problems are implemented. For all sampling frequencies both single step and N-point multi step prediction results obtained using LAM and LLM with regularization methods perform better than IIR prediction filters for the selected sample patients. Moreover, since the simple LAM model performs as well as the more complicated LLM models in our patient sample, its use for non-linear prediction is recommended.",
keywords = "Breathing pattern, Embedding dimension, Infinite response prediction filters, Lung cancer, Non-linear dynamics, Non-linear prediction, Prediction, Time delay",
author = "Tolakanahalli, {R. P.} and Tewatia, {D. K.} and Tome, {Wolfgang A.}",
year = "2015",
month = "5",
day = "1",
doi = "10.1016/j.ejmp.2015.01.018",
language = "English (US)",
volume = "31",
pages = "257--265",
journal = "Physica Medica",
issn = "1120-1797",
publisher = "Associazione Italiana di Fisica Medica",
number = "3",

}

TY - JOUR

T1 - Time series prediction of lung cancer patients' breathing pattern based on nonlinear dynamics

AU - Tolakanahalli, R. P.

AU - Tewatia, D. K.

AU - Tome, Wolfgang A.

PY - 2015/5/1

Y1 - 2015/5/1

N2 - This study focuses on predicting breathing pattern, which is crucial to deal with system latency in the treatments of moving lung tumors. Predicting respiratory motion in real-time is challenging, due to the inherent chaotic nature of breathing patterns, i.e. sensitive dependence on initial conditions. In this work, nonlinear prediction methods are used to predict the short-term evolution of the respiratory system for 62 patients, whose breathing time series was acquired using respiratory position management (RPM) system. Single step and N-point multi step prediction are performed for sampling rates of 5Hz and 10Hz. We compare the employed non-linear prediction methods with respect to prediction accuracy to Adaptive Infinite Impulse Response (IIR) prediction filters. A Local Average Model (LAM) and local linear models (LLMs) combined with a set of linear regularization techniques to solve ill-posed regression problems are implemented. For all sampling frequencies both single step and N-point multi step prediction results obtained using LAM and LLM with regularization methods perform better than IIR prediction filters for the selected sample patients. Moreover, since the simple LAM model performs as well as the more complicated LLM models in our patient sample, its use for non-linear prediction is recommended.

AB - This study focuses on predicting breathing pattern, which is crucial to deal with system latency in the treatments of moving lung tumors. Predicting respiratory motion in real-time is challenging, due to the inherent chaotic nature of breathing patterns, i.e. sensitive dependence on initial conditions. In this work, nonlinear prediction methods are used to predict the short-term evolution of the respiratory system for 62 patients, whose breathing time series was acquired using respiratory position management (RPM) system. Single step and N-point multi step prediction are performed for sampling rates of 5Hz and 10Hz. We compare the employed non-linear prediction methods with respect to prediction accuracy to Adaptive Infinite Impulse Response (IIR) prediction filters. A Local Average Model (LAM) and local linear models (LLMs) combined with a set of linear regularization techniques to solve ill-posed regression problems are implemented. For all sampling frequencies both single step and N-point multi step prediction results obtained using LAM and LLM with regularization methods perform better than IIR prediction filters for the selected sample patients. Moreover, since the simple LAM model performs as well as the more complicated LLM models in our patient sample, its use for non-linear prediction is recommended.

KW - Breathing pattern

KW - Embedding dimension

KW - Infinite response prediction filters

KW - Lung cancer

KW - Non-linear dynamics

KW - Non-linear prediction

KW - Prediction

KW - Time delay

UR - http://www.scopus.com/inward/record.url?scp=84925108870&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84925108870&partnerID=8YFLogxK

U2 - 10.1016/j.ejmp.2015.01.018

DO - 10.1016/j.ejmp.2015.01.018

M3 - Article

VL - 31

SP - 257

EP - 265

JO - Physica Medica

JF - Physica Medica

SN - 1120-1797

IS - 3

ER -