Appendix A - Prediction and Classification of Respiratory Motion - page 134

HEKF Hybrid extended Kalman filter

MC-IMME Multi-channel interacting multiple model estimator

IMME Interacting multiple model estimator

GMM Gaussian mixture model

EM Expectation-maximization

MM Multiple model

AMM Autonomous multiple models

CMM Cooperating multiple models

VSMM Variable structure multi-models

HEKF Hybrid implementation based on EKF

NRMSE Normalized root mean squared error

RNN Recurrent neural network

BPTT Back-propagation-through-time

RTRL Real-time recurrent learning

MLP Multilayer perceptron

RMLP Recurrent multilayer perceptron

CNN Customized prediction with multiple patient interaction using NN

AMV Autocorrelation MAX value

ADT Autocorrelation delay time

ACC Acceleration variance value

VEL Velocity variance value

BRF Breathing frequency

FTP Max power of Fourier transform

PCA Principal component analysis coefficient

MLR Multiple linear regression coefficient

STD Standard deviation of time series data

MLE Maximum likelihood estimates

KDE Kernel density estimation

O-ANN Optimized adaptive neural network

ALP Adaptive linear prediction

ROC Receiver operating characteristics

AUC Area under the ROC curve

AIC Akaike information criterion

BIC Bayesian information criterion

ITV Internal target volume

3D-CRT Three-dimensional conformal radiotherapy

ECG Electrocardiogram

TPR True positive rate

TP True positive

FPR False positive rate

FP False positive

TN True negative

FN False negative