Categorizing automatically generated physiological data based on industry guidelines

• US 8,690,789 B2
• Filed: 05/09/2011
• Issued: 04/08/2014
• Est. Priority Date: 08/08/2006
• Status: Active Grant

• Current Assignee: 3M Innovative Properties Company (3M Company)
• Original Assignee: 3M Innovative Properties Company (3M Company)
• Inventors: Watrous, Raymond L.
• Primary Examiner(s): BEHRINGER, LUTHER G
• Application Number:
  US13/103,750
• Publication Number:
  US 20110208080A1
• Time in Patent Office: 1,065 Days
• Field of Search: 600/528
• US Class Current: 600/528
• CPC Class Codes: A61B 7/04 Electric stethoscopes microphones, acoustic transducers therefor H04R
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 70/20 relating to practices or guidelines

Abstract

Methods and systems for mapping a physiological signal into clinical guideline parameters are disclosed. A physiological signal having a characteristic that may represent an anomaly is received and mapped to a clinical guideline condition space. Probabilities are determined that the mapped signal with which the anomaly may be associated represents a first clinical guideline condition corresponding to a referral indication or a second clinical guideline condition corresponding to an absence of the referral indication. The determined probability is presented and a referral decision is made responsive to the determined probability that the anomaly is associated with the first clinical guideline condition.

Claims

1. A method for assisting in a referral indication of heart murmurs, the method comprising the steps of:

receiving an acoustic signal representing heart sounds;

parsing the received acoustic signal into predetermined physiological events including a plurality of systolic intervals and a plurality of diastolic intervals;

computing a first normalized mid-range energy representative of a median systolic interval and a second normalized mid-range energy representative of a median diastolic interval, respectively, the first and second normalized mid-range energies computed from the parsed acoustic signal;

characterizing the median systolic interval based on the first normalized mid-range energy by computing a focal point of energy using a maximum first normalized mid-range energy and systolic interval temporal center of a normalized mid-range energy concentration to generate a model of the median systolic interval;

determining probabilities relative to a first condition associated with the referral indication and a second condition associated with an absence of the referral indication based on the characterized median systolic interval; and

presenting the first normalized mid-range energy, the second normalized mid-range energy and the determined probabilities,

whereby the presented first normalized mid-range energy, the second normalized mid-range energy and the determined probabilities may be used to determine whether the referral is indicated.

2. The method according to claim 1, the method including the steps of:

characterizing the median diastolic interval based on the second normalized mid-range energy;

determining further probabilities relative to a diastolic murmur condition and a non-diastolic murmur condition; and

the step of presenting further includes presenting the further determined probabilities.

3. The method according to claim 1, wherein the first and second normalized mid-range energies are each between 150 and 600 Hz.

4. The method according to claim 1, wherein the first and second normalized mid-range energies are converted to respective murmur grades and the step of presenting the first and second normalized mid-range energies presents the respective murmur grades.

5. The method according to claim 1, wherein the first condition includes systolic murmurs selected from the group consisting of a holo-systolic murmur, a late-systolic murmur and a mid-systolic murmur equal or greater than a murmur grade 3.

6. The method according to claim 1, wherein the step of presenting the determined probabilities presents derived features relative to a probability surface, the probability surface including a first region representing the first condition and a second region representing the second condition.

7. The method according to claim 6, wherein the probability surface includes a predetermined threshold between the first region and the second region, the predetermined threshold selected according to a predetermined sensitivity and a predetermined specificity.

8. The method according to claim 7, wherein the probability surface is presented as a function of a duration of the median systolic interval and the first normalized mid-range energy associated with the median systolic interval.

9. The method according to claim 1, the step of computing the first normalized mid-range energy and second normalized mid-range energy includes the steps of:

dividing each of the median systolic interval and median diastolic intervals into a plurality of subintervals; and

calculating the first normalized mid-range energy and second normalized mid-range energy for each of the subintervals of the respective median systolic interval and median diastolic interval.

10. The method according to claim 9 wherein:

the step of dividing each of the median systolic interval and median diastolic intervals into a plurality of subintervals includes the steps of;

subdividing the median systolic interval into at least thirty intervals;

subdividing the median diastolic interval into at least thirty intervals, and

the step of presenting the first and second normalized mid-range energies includes the step of graphically displaying the calculated normalized mid-range energy for each of the plurality of sub-intervals.

11. The method according to claim 1, including the step of detecting heart murmurs using the parsed audio signal and the first normalized mid-range energy and second normalized mid-range energy.