Blood vessel mechanical signal analysis
Disclosed herein is a framework for facilitating patient signal analysis. In accordance with one aspect, the framework receives signal data including mechanical signal data, wherein the mechanical signal data is generated in response to mechanical contraction of blood vessels. A region of interest is segmented from the mechanical signal data. One or more mechanical signal ratios may be determined based on parameters extracted from the segmented region of interest to characterize waveform changes. A report may then be generated based at least in part on the one or more mechanical signal ratios.
1. A system for patient signal analysis, comprising:
a sensor system including sensors that non-invasively acquire at least first and second types of mechanical signal data from a patient, wherein the at least first and second types of mechanical signal data is generated in response to contraction of blood vessels; and
a computer system communicatively coupled to the sensor system, wherein the sensor system is configured to continuously and wirelessly transmits the first and second types of mechanical signal data to the computer system, wherein the computer system includes
a non-transitory memory device for storing computer readable program code, and
a processor in communication with the memory device, the processor being operative with the computer readable program code to perform steps including
segmenting a first region of interest from each of the first and second types of mechanical signal data into first and second different portions by using a predetermined percentage of first and second maximum amplitudes of the mechanical signal data, wherein the first region of interest corresponds to a cardiac cycle,
determining first and second mechanical signal ratios based on first and second parameters extracted from the first and second portions of the first and second types of mechanical signal data, wherein the first and second mechanical signal ratios characterize waveform changes, wherein the first mechanical signal ratio comprises a time integration ratio of an integral of time domain magnitudes of the first portion to an integral of time domain magnitudes of the second portion,
determining a third mechanical signal ratio comprising a frequency energy integration ratio that compares integrals of frequency spectral magnitudes of first and second portions of a second region of interest of the mechanical signal data in a frequency domain,
integrating, via an artificial neural network, at least the first, second and third mechanical signal ratios to generate output results for detecting a cardiac pathology, and
generating a report based at least in part on the output results.
2. The system of claim 1 wherein the sensor system is configured to be removably attachable to a wrist.
3. The system of claim 1 wherein the one or more sensors comprise a pulse vibration sensor and the mechanical signal data comprises vibration signal data.
4. The system of claim 1 wherein the one or more sensors comprise an acoustic sensor and the mechanical signal data comprises acoustic signal data.
5. The system of claim 1 wherein the one or more sensors comprise an optical sensor and the mechanical signal data comprises oximetric signal data.
6. The system of claim 1 wherein the sensor system further comprises an indicator that provides warning of a detected cardiac pathology.
7. A method of patient signal analysis, comprising:
continuously and wirelessly receiving, by a processor device from a sensor system, patient signal data including at least first and second types of mechanical signal data generated in response to contraction of blood vessels;
segmenting, by the processor device, a first region of interest from each of the first and second types of mechanical signal data into first and second different portions by using a predetermined percentage of first and second maximum amplitudes of the mechanical signal data, wherein the first region of interest corresponds to a cardiac cycle;
determining, by the processor device, first and second mechanical signal ratios based on first and second parameters extracted from the first and second portions of the first and second types of mechanical signal data, wherein the first and second mechanical signal ratios characterize waveform changes, wherein the first mechanical signal ratio comprises a time integration ratio of an integral of time domain magnitudes of the first portion to an integral of time domain magnitudes of the second portion;
determining, by the processor device, a third mechanical signal ratio comprising a frequency energy integration ratio that compares integrals of frequency spectral magnitudes of first and second portions of a second region of interest of the mechanical signal data in a frequency domain;
integrating, via an artificial neural network, at least the first, second and third mechanical signal ratios to generate output results for detecting a cardiac pathology; and
generating, by the processor device, a report based at least in part on the output results.
8. The method of claim 7 wherein determining the first and second mechanical signal ratios comprises determining one or more ratios of a first value to a second value, wherein the first value is determined based on a first parameter extracted from the first portion and the second value is determined based on a second parameter extracted from the second portion.
9. The method of claim 7 further comprising extracting time durations from the first region of interest to generate the first mechanical signal ratio.
10. The method of claim 7 wherein the frequency energy integration ratio comprises a unipolar ratio that compares parameters extracted from portions in a same cardiac cycle.
11. The method of claim 7 further comprising:
extracting frequency peaks from the first and second portions of the second region of interest; and
determining a fourth mechanical signal ratio of a first value to a second value, wherein the first value is derived based on the frequency peak from the first portion of the second region of interest and the second value is derived based on the frequency peak from the second portion of the second region of interest.
12. The method of claim 11 wherein determining the fourth mechanical signal ratio comprises determining a dominant frequency ratio that compares dominant peak frequency values of the first and second portions of the second region of interest.
13. The method of claim 7 wherein determining the first mechanical signal ratio comprises determining a unipolar ratio that compares parameters extracted from portions in a same cardiac cycle.
14. The method of claim 7 wherein determining the first mechanical signal ratio comprises determining a bipolar ratio that compares parameters extracted from portions in different cardiac cycles.
15. The method of claim 7 wherein determining the second mechanical signal ratio comprises determining a mutual ratio that compares parameters extracted from same portions in different cardiac cycles.
16. The method of claim 7 wherein determining the second mechanical signal ratio comprises determining a cross ratio that compares parameters extracted from different portions in different cardiac cycles.
17. The method of claim 7 wherein determining the second mechanical signal ratio comprises determining a time ratio that compares time durations of the first and second portions of the first region of interest.
18. The method of claim 7 wherein the frequency energy integration ratio comprises a bipolar ratio that compares parameters extracted from portions in different cardiac cycles.
19. A non-transitory computer readable medium embodying a program of instructions executable by machine to perform steps for patient signal analysis, the steps comprising:
continuously and wirelessly receiving, by a processor device from a sensor system, patient signal data including at least first and second types of mechanical signal data generated in response to contraction of blood vessels;
segmenting, by the processor device, a first region of interest from each of the first and second types of mechanical signal data into first and second different portions by using a predetermined percentage of first and second maximum amplitudes of the mechanical signal data, wherein the first region of interest corresponds to a cardiac cycle;
determining, by the processor device, first and second mechanical signal ratios based on first and second parameters extracted from the first and second portions of the first and second types of mechanical signal data, wherein the first and second mechanical signal ratios characterize waveform changes, wherein the first mechanical signal ratio comprises a time integration ratio of an integral of time domain magnitudes of the first portion to an integral of time domain magnitudes of the second portion;
determining, by the processor device, a third mechanical signal ratio comprising a frequency energy integration ratio that compares integrals of frequency spectral magnitudes of first and second portions of a second region of interest of the mechanical signal data in a frequency domain;
integrating, via an artificial neural network, at least the first, second and third mechanical signal ratios to generate output results for detecting a cardiac pathology; and
generating, by the processor device, a report based at least in part on the output results.