Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11190166-B2
Publication Date
2021-11-30
Expiration Date
Abstract
The present invention provides a system and method for representing quasi-periodic (“qp”) waveforms comprising, representing a plurality of limited decompositions of the qp waveform, wherein each decomposition includes a first and second amplitude value and at least one time value. In some embodiments, each of the decompositions is phase adjusted such that the arithmetic sum of the plurality of limited decompositions reconstructs the qp waveform. These decompositions are stored into a data structure having a plurality of attributes. Optionally, these attributes are used to reconstruct the qp waveform, or patterns or features of the qp wave can be determined by using various pattern-recognition techniques. Some embodiments provide a system that uses software, embedded hardware or firmware to carry out the above-described method. Some embodiments use a computer-readable medium to store the data structure and/or instructions to execute the method.
Core Innovation
The invention relates to analyzing a quasi-periodic waveform that includes a cardiac signal by obtaining a digitized signal with a series of digital values, storing the digitized values, and generating a series of states defined by phase relationships between a plurality of frequency components of the cardiac signal. The series of states is generated based on the stored series of digital values rather than by directly using raw time-domain morphology, so that phase relationships among frequency components become the primary representation.
The invention automatically generates and outputs representations of segments of the cardiac signal based on the series of states. The segment representations include graphical representations and matrix-table representations, and representations based on points marked by fractional-phase transitions across frequency components.
In additional embodiments, phase sequences use quarter-phase and fractional-phase representations based on detected zero crossings or peaks, with phase labels associated with state elements. Fractional-phase representations and resulting state vectors are linked horizontally and vertically into a mapped state space or vector space for pattern recognition using Hidden Markov Models and neural networks, with optional reconstruction of the quasi-periodic waveform from stored objects or components.
Claims Coverage
The provided material includes three independent claims: clm-00001 (apparatus), clm-00004 (computer-implemented method), and clm-00019 (non-transitory computer-readable medium). Across these independent claims, the coverage centers on deriving a series of phase-relationship-based states from a quasi-periodic cardiac signal’s frequency components and automatically generating and outputting segment representations based on the state series.
State generator based on phase relationships between frequency components
Generating a series of states defined by phase relationships between a plurality of frequency components of the cardiac signal based on the stored series of digital values.
Automatic segment generator outputting representations based on states
Automatically outputting representations of segments of the cardiac signal based on the series of states.
Digitized quasi-periodic cardiac signal obtainment and storage
Obtaining a digitized signal having a series of digital values of a quasi-periodic waveform, wherein the quasi-periodic waveform includes a cardiac signal; and storing the series of digital values in the storage unit.
Each independent claim requires a digitized quasi-periodic cardiac signal representation obtained and stored, generation of a series of states defined by phase relationships among a plurality of frequency components, and automatic generation and output of representations of cardiac signal segments based on the generated series of states.
Stated Advantages
Not explicitly described in patent.
Documented Applications
ECG analysis/segmentation of a cardiac signal, including automatic output of representations of segments based on phase-relationship-defined states.
Pattern recognition in cardiac signal state sequences using Hidden Markov Models and neural networks.
Classification/examples comparing acute myocardial infarction versus normal sinus rhythm and detecting QRS boundaries.
Interested in licensing this patent?