SYSTEM AND METHOD FOR PAIN MONITORING USING A MULTIDIMENSIONAL ANALYSIS OF PHYSIOLOGICAL SIGNALS

• US 20130310660A1
• Filed: 07/18/2013
• Published: 11/21/2013
• Est. Priority Date: 11/14/2007
• Status: Active Grant

• Current Assignee: Medasense Biometrics Ltd.
• Original Assignee: Medasense Biometrics Ltd.
• Inventors: Zuckerman-Stark, Galit, Kliger, Mark
• Granted Patent: US 9,498,138 B2
• Time in Patent Office: Days
• US Class Current: 600/301
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 3/112 for measuring diameter of pupils
• CPC Class Codes: A61B 5/02 Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
• CPC Class Codes: A61B 5/02055 Simultaneously evaluating both cardiovascular condition and temperature
• CPC Class Codes: A61B 5/021 Measuring pressure in heart or blood vessels A61B5/0205 takes precedence
• CPC Class Codes: A61B 5/02405 Determining heart rate variability
• CPC Class Codes: A61B 5/02416 using photoplethysmograph signals, e.g. generated by infrared radiation A61B5/14552 takes precedence
• CPC Class Codes: A61B 5/0261 using optical means, e.g. infrared light
• CPC Class Codes: A61B 5/0295 using plethysmography, i.e. measuring the variations in the volume of a body part as modified by the circulation of blood therethrough, e.g. impedance plethysmography
• CPC Class Codes: A61B 5/0533 Measuring galvanic skin response
• CPC Class Codes: A61B 5/0836 Measuring rate of CO2 production
• CPC Class Codes: A61B 5/24 Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
• CPC Class Codes: A61B 5/318 Heart-related electrical modalities, e.g. electrocardiography [ECG]
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/392 Detecting gastrointestinal contractions
• CPC Class Codes: A61B 5/398 Electrooculography [EOG], e.g. detecting nystagmus; Electroretinography [ERG]
• CPC Class Codes: A61B 5/4821 Determining level or depth of anaesthesia A61B5/1106 takes precedence
• CPC Class Codes: A61B 5/4824 Touch or pain perception evaluation A61B5/4821 takes precedence
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• CPC Class Codes: A61B 5/7264 : Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7267 : involving training the classification device
• CPC Class Codes: A61B 5/7275 : Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: A61B 5/7282 : Event detection, e.g. detecting unique waveforms indicative of a medical condition A61B5/4094, A61B5/4818, A61B5/349, A61B5/0823 take precedence
• CPC Class Codes: G06N 20/00 : Machine learning
• CPC Class Codes: G16H 40/63 : for local operation
• CPC Class Codes: G16H 50/20 : for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 : for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16H 50/70 : for mining of medical data, e.g. analysing previous cases of other patients

Abstract

The present invention is for a method and system for pain classification and monitoring optionally in a subject that is an awake, semi-awake or sedated.

Claims

1. A method for detecting and classifying the pain status of a patient by analyzing a plurality of physiological signals, said method comprises:

a. Signal acquisition for acquiring said plurality of physiological signals; and

b. pre-processing said acquired plurality of physiological signals to improve signal quality comprising at least one or more chosen from the group consisting of synchronization, noise filtering, artifact reduction, therein forming a plurality of pre-processed physiological signals; and

c. processing said pre-processed plurality of physiological signals, said processing comprising;

i. feature extraction from at least three or more physiological signals including PPG, GSR and at least one or more physiological signals chosen from the group consisting;

of skin temperature, ECG, Respiration, EMG, and EEG/FEMG to facilitate detection of pain in a patient; and

forming a first vector comprising a set of extracted features; and

ii. transforming said first vector to a second vector wherein pain detection is performed based on said second vector and wherein said transformation comprises normalization, feature selection and dimensionality reduction techniques; and

iii. detecting the pain status of a patient by applying a classification function to classify said second vector into at least two classes of pain.

2. The method of claim 1 wherein said pain detection further comprises:

d. communicating said detected pain status of said patient to at least one or more chosen from the group consisting of a higher processing center, person, caregiver, call center and any combination thereof.

3. The method of claim 1 wherein said first vector comprises a plurality of features including parameters extracted from said PPG and GSR physiological signals including:

a. PPG features chosen from the group consisting of;

i. PPG Peak (P) amplitude, Trough (T) amplitude, mean PPG Peak (P) amplitude, and std of PPG Peak (P) amplitude, mean Trough (T) amplitude, and std of Trough (T) amplitude; and

ii. PPG peak to peak time intervals, PPG peak to peak interval mean and PPG peak to peak interval std; and

iii. power spectrum of the PPG peak to peak intervals;

VLF Power. LF Power and HF Power; and

b. GSR features chosen from the group consisting of;

i. GSR amplitude, GSR mean amplitude and GSR amplitude std; and

ii. GSR Peak (P) amplitude, mean Peak (P) amplitude and Peak (P) amplitude std; and

iii. GSR peak to peak time intervals, mean GSR peak to peak time interval; and

GSR peak to peak time intervals std; and

iv. Phasic EDA;

amplitude, mean amplitude and std of amplitude.

4. The method of claim 3 wherein said first vector further comprises a group of features extracted from one of said physiological signals selected from the group consisting of skin temperature, ECG, Respiration, EMG, and EEG/FEMG.

5. The method of claim 3 wherein said first vector further comprises a group of features extracted from two of said physiological signals selected from the group consisting of skin temperature, ECG, Respiration, EMG, and EEG/FEMG.

6. The method of claim 3 wherein said first vector further comprises a group of features extracted from three of said physiological signals selected from the group consisting of skin temperature, ECG, Respiration, EMG, and EEG/FEMG.

7. The method of claim 3 wherein said first vector further comprises a group of features extracted from four of said physiological signals selected from the group consisting of skin temperature, ECG, Respiration, EMG, and EEG/FEMG.

8. The method of claim 3 wherein said first vector further comprises a group of features extracted from said physiological signals essentially consisting of PPG, GSR, skin temperature. ECG, Respiration, EMG, and EEG/FEMG.

9. The method of claim 3 wherein said extracted features are chosen form at least one or more of the feature groups corresponding to a physiological signal chosen from the group consisting of skin temperature, ECG, Respiration, EMG, and EEG/FEMG, said feature groups comprising:

a. Skin temperature features chosen from the group of consisting;

i. Temperature amplitude, mean amplitude and std of amplitude; and

ii. Temp Peak (P) amplitude, mean amplitude and std of amplitude; and

iii. Temperature peak to peak time intervals, mean and std (variability) of interval; and

b. ECG features chosen from the group of consisting;

i. ECG-PPG PTT Pulse Transition time; and

ii. ECG R to R time intervals, mean and std (variability) of intervals; and

iii. Power of VLF, LF and HF frequency bands of power spectrum of the ECG R to R intervals (heart rate variability); and

c. Respiration features chosen from the group of consisting of;

i. Upper peak amplitude, mean amplitude and STD of amplitude; and

ii. Respiratory rate, mean rate and std rate; and

d. EMG features chosen from the group of consisting of;

i. Power of the frequency bands of power spectrum of EMG signal; and

ii. EMG Power Spectrum Mean frequency; and

iii. EMG Power Spectrum Highest Peak Frequency; and

e. EEG/FEMG features chosen from the group of consisting of;

i. Power of the alpha (α

), beta (β

), gamma (γ

), delta (δ

), theta (θ

) frequency bands of power spectrum of EEG/FEMG signal; and

ii. EMG Power Spectrum Mean frequency; and

iii. EMG Power Spectral edge frequency; and

iv. Coherence between 2 or more EEG/FEMG channels; and

f. any combination thereof.

10. The method of claim 1 wherein said signal acquisition facilitates acquiring physiological signals with a plurality of sensors chosen from the group consisting of:

ECG, PPG, blood pressure, respiration, internal body temperature, skin temperature, EOG, pupil diameter monitoring, GSR, EEG/FEMG, EMG, EGG, LDV, capnograph and accelerometer, and wherein said physiological signals are pre-processed and processed facilitating said detection and classification of pain status.

11. The method of claim 1 further comprising obtaining and processing a priori data for facilitating said detection and classification of pain status, wherein said a priori data is selected from the group consisting of:

patient associated data, historical data, data from the group consisting of data supplied by the physician'"'"'s, environmental parameters, patient parameters, disease, stimulus and medicament or any combination thereof; and

wherein said second vector further comprises said a priori data.

12. The method of claim 1 wherein said pain classification is classified into groups selected form the group consisting of:

at least two classes, at least three classes, pain/no-pain states, graduated scale, scale of 1 to 10, a scale of 1-100, subjective pain scaling schemes, subjective scoring schemes, Visual Analog Scale (VAS), Numeric Pain Scale (NPS) and verbal scale.

13. The method according to claim 1 wherein said dimensionality reducing transformation comprises a linear or non-linear transformation is selected from the group consisting of:

Multi Dimensional Scaling (MDS), Principal Component Analysis (PCA), Sparse PCA (SPCA), Fisher Linear Discriminant Analysis (FLDA), Sparse FLDA (SFLDA), Kernel PCA (KPCA), ISOMAP, Locally Linear Embedding (LLE), Laplacian Eigenmaps, Diffusion Maps, Hessian Eigenmaps, Independent Component Analysis (ICA), Factor analysis (FA), Hierarchical Feature selection and dimensionality reduction (HDR), Sure Independence Screening (SIS), Fisher score ranks, t-test rank, Mann-Whitney U-test or any combination thereof.

14. The method of according to claim 1, wherein said classification function is a linear or non-linear transformation selected from a group consisting of:

Nearest Shrunken Centroids (NSC). Classification and Regression Trees (CART), ID3. C4.5, Multivariate Additive regression splines (MARS), Multiple additive regression trees (MART), Nearest Centroid (NC), Shrunken Centroid Regularized Linear Discriminate and Analysis (SCRLDA), Random Forest, Boosting, Bagging Classifier, AdaBoost, RealAdaBoost, LPBoost, TotalBoost, BrownBoost, MadaBoost, LogitBoost, GentleBoost, RobustBoost, Support Vector Machine (SVM), kernelized SVM, Linear classifier, Quadratic Discriminant Analysis (QDA) classifier, Naï

ve Bayes Classifier and Generalized Likelihood Ratio Test (GLRT) classifier with plug-in parametric or non-parametric class conditional density estimation, k-nearest neighbor, Radial Base Function (RBF) classifier, Multilayer Perceptron classifier, Bayesian Network (BN) classifier, multi-class classifier adapted from binary classifier with one-vs-one majority voting, one-vs-rest, Error Correcting Output Codes, hierarchical multi-class classification, Committee of classifiers or any combination thereof.

15. The method according to claim 1 wherein said feature selection and dimensionality reduction is provided by Hierarchical Dimensionality Reduction (HDR);

and wherein said classification is provided by a Random Forest classifier.

16. The method according to claim 1 wherein said, feature selection and dimensionality reduction is provided by Fisher Score rank and SFLDA;

and wherein said classification is provided by a RealAdaboost classifier within a Boosting framework.

17. The method of claim 1 wherein said classification is adapted for pain experienced with a particular disease, stimulus or medicament.

18. The method of claim 1 wherein said status of pain is adapted for a patient in various states of consciences chosen from the group consisting of:

sedated, partially sedate and awake, semi-awake.

19. A system for detecting and classifying the pain status of a patient by analyzing a plurality of physiological signals comprising:

a. a signal acquisition module comprising a plurality of sensors and/or transducers for measuring and/or obtaining said at least three or more physiological signals and a priori data from a subject; and

b. a processing module for processing said physiological signals, comprising;

i. pre-processing said acquired plurality of physiological signals to improve signal quality comprising at least one or more tools chosen from the group consisting of synchronization, noise filtering, artifact reduction therein forming a pre-processed plurality of physiological signals; and

ii. processing said pre-processed plurality of physiological signals comprising;

iii. feature extraction from at least three or more physiological signals including PPG, GSR and at least one or more physiological signals chosen from the group consisting;

of skin temperature, ECG, Respiration, EMG, and EEG/FEMG to facilitate detection of pain in a patient; and

forming a first vector comprising a set of extracted features; and

iv. transforming said first vector to a second vector wherein pain detection is performed based on said second vector and wherein said transformation comprises normalization and feature selection and dimensionality reduction techniques; and

v. detecting the pain status of a patient by applying a classification function to classifying said second vector into at least two classes of pain; and

c. a communicating module for communicating said detected pain status of said patient to at least one or more chosen from the group consisting of a higher processing center, person, caregiver, call center and any combination thereof.

20. The system of claim 19 further comprising a display module for displaying said detected and classified pain stimulus.