System and method for pain monitoring using a multidimensional analysis of physiological signals
Inventors
Zuckerman-Stark, Galit • Kliger, Mark
Assignees
Publication Number
US-10743778-B2
Publication Date
2020-08-18
Expiration Date
2028-11-13
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
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.
Core Innovation
The invention provides a system and method for pain monitoring by performing a multidimensional analysis of a plurality of physiological signals to classify and identify pain levels in individuals exhibiting various states of consciousness, including awake, semi-awake, sedated, and unconscious states. The method involves acquiring multiple physiological signals, extracting a great plurality of features (GPF), normalizing and reducing dimensionality of these features, and then classifying the pain level using machine learning classifiers.
The problem being solved addresses the limitations of previous pain monitoring and Depth of Anesthesia (DOA) systems which typically rely on single physiological parameters or are limited to unconscious patients under anesthesia. Prior art used isolated physiological signals such as skin conductance, EEG, ECG, or blood pressure, but medical studies showed that combining parameters from different physiological signals significantly improves pain-no pain classification. Furthermore, existing systems often did not provide reliable monitoring across various consciousness states, nor comprehensive multidimensional analysis to detect and classify pain effectively in awake or sedated patients.
Claims Coverage
The patent includes multiple independent claims focused on a method and system for monitoring pain using multiple physiological signals and advanced classification algorithms.
Method for monitoring pain using multidimensional physiological signals
Obtaining at least two physiological signals including Galvanic Skin Response and at least one from a defined group such as blood volume change, PPG, ECG, respiration, temperature, EEG, EMG, and others; processing these signals to improve quality; generating a first feature vector comprising at least three features including specific GSR features; transforming this vector through normalization; and monitoring pain status by applying a classification algorithm that classifies into a graduated scale with at least three pain levels, wherein the classification algorithm comprises an ensemble of classification and regression trees, and the patient can be unconscious.
Inclusion of extensive physiological features for pain classification
Using a comprehensive and specific set of features extracted from multiple physiological signals such as PPG features including maximum rate point and intervals; ECG features including RR intervals and heart rate variability in frequency bands; temperature features; respiratory features; EEG/EMG features including power in multiple frequency bands and entropy measures; and others as detailed in the claims, integrated to improve pain state detection.
Communication and incorporation of a priori data
Optionally obtaining and processing a priori data related to patient or environmental parameters such as disease, stimulus, and medicaments to enhance pain detection and classification accuracy, and communicating monitored pain status to external entities like caregivers or higher processing centers.
System for pain monitoring with signal acquisition, processing, and communication modules
A system comprising a signal acquisition module with multiple sensors/transducers measuring at least two physiological signals including GSR and others from a specified group; a processing module performing signal quality improvement, feature extraction, transformation, and classification into at least three pain levels using ensemble methods like random forest or boosting frameworks; optionally comprising a display module and communication module for output.
The claims cover inventive methods and systems that obtain multiple physiological signals, extract a defined set of features including GSR-related ones, implement normalization and dimensionality reduction, and apply advanced classification algorithms such as ensemble methods to monitor and classify pain levels, applicable to patients in various consciousness states including unconscious.
Stated Advantages
Significant improvement in pain and no-pain classification performance by using combinations of multiple physiological parameters instead of single signals alone.
Capability to monitor pain in various states of consciousness including awake, semi-awake, sedated, and unconscious patients.
Robust multidimensional analysis combining a large set of physiological features providing a more accurate and sensitive pain classification.
Flexibility to include a priori data to enhance classification efficiency.
Modular system design with signal acquisition, processing, classification, communication, and display modules allowing remote or local configurations.
Documented Applications
Monitoring and classification of pain levels in patients across conscious states, including sedation and anesthesia.
Clinical pain monitoring during procedures involving anesthesia or sedation.
Remote pain monitoring systems for telemedicine applications involving communication of pain status to caregivers or medical centers.
Interested in licensing this patent?