Hierarchial analysis for processing brain stem signals to define a prominent wave

• US 5,003,986 A
• Filed: 11/17/1988
• Issued: 04/02/1991
• Est. Priority Date: 11/17/1988
• Status: Expired due to Term

• Current Assignee: Electrophysiology Laboratories of Dallas Incorporated, Kenneth D. Pool Jr
• Original Assignee: Electrophysiology Laboratories of Dallas Incorporated, Kenneth D. Pool Jr
• Inventors: Pool, Kenneth D. Jr., Finitzo, Terese
• Primary Examiner(s): Jaworski, Francis
• Application Number:
  US07/272,714
• Time in Patent Office: 866 Days
• Field of Search: 128/731-733, 128/680, 364/413.02-413.05
• US Class Current: 600/544
• CPC Class Codes: A61B 5/121 evaluating hearing capacity
• CPC Class Codes: A61B 5/38 Acoustic or auditory stimuli

Abstract

A signal waveform analyzation technique which filters digital data to define primary peaks. Unfiltered data of the primary peaks is then employed to define secondary peaks associated with the primary peak. Amplitude and time characteristic data associated with the secondary peaks is then employed to determine reproducibility of secondary peaks between the pair of traces. A hierarchial analysis is utilized to further analyze the reproducible pairs of secondary peaks to identify a prominent wave. Depending on the success of a first pass, the digital data is filtered with a different bandpass, whereupon the analyzation steps are again repeated in an attempt to identify a prominent wave.

Claims

1. A method for analyzing signal waveform trace data to identify a prominent wave, comprising the steps of:

analyzing digital signals of related first and second waveform traces to define primary peaks of said first and second waveform traces which satisfy a predetermined criteria, each said primary peak having associated preceding and succeeding troughs;

if a primary peak is not determined by initial analyzation of said data signals, filtering said digital signals with a predefined bandpass and again analyzing the filtered digital signals until one said primary peak is found in each said trace;

analyzing said primary peaks to determine secondary peaks;

analyzing said secondary peaks to characterize said secondary peaks according to predefined criteria;

determining reproducibility of said secondary peaks between said first and second waveform traces; and

identifying said prominent wave by employing reproducibility criteria for the secondary peaks of said first and second waveform traces.

2. The method of claim 1, further including analyzing said digital signals between two predefined time periods of said first and second signal waveform traces.

3. The method of claim 2, further including analyzing data between a latency period of about 5.2-10.5 milliseconds for identifying a Wave V of a brain stem signal waveform.

4. The method of claim 1, wherein said primary peak and said preceding and succeeding troughs are determined by analyzing said digital signals for slope characterization as measured over a number of digital signals.

5. The method of claim 1, further including identifying said primary peak when an amplitude characteristic of a brain stem signal waveform exceeds about 0.06 microvolts.

6. The method of claim 1, further including analyzing pairs of alternately arranged digital signals to determine said secondary peaks.

7. The method of claim 6, further including analyzing only a downward slope of a primary peak to a succeeding trough digital signals to define amplitude characteristics of said sample pairs by carrying out a subtraction between said digital signal pairs, and comparing a result thereof with a predefined number.

8. The method of claim 1, further including, for each said secondary peak, analyzing digital signals in a reverse direction from a secondary peak to define a preceding valley associated with said secondary peak.

9. The method of claim 1, further including calculating an amplitude characteristic of each said secondary peak, and defining those secondary peaks whose amplitudes exceed a predetermined amplitude as candidates for identifying said prominent wave.

10. The method of claim 9, further including defining said prominent wave candidates as said secondary peaks with amplitudes exceeding 60% of an associated primary peak amplitude, when analyzing brain stem waveforms.

11. The method of claim 1, further including calculating an amplitude characteristic of each said secondary peak utilizing an amplitude from a peak to an associated succeeding valley.

12. The method of claim 11, further including defining said prominent wave by comparing reproducible secondary peaks having a largest secondary peak to succeeding valley amplitude.

13. The method of claim 1, further including defining said prominent wave by comparing reproducible secondary peaks, one said secondary peak of the reproducible pair having a largest secondary peak to succeeding valley amplitude, and the other secondary peak of the reproducible pair having a second largest secondary peak to succeeding valley amplitude.

14. The method of claim 13, wherein said secondary peak is selected as being reproducible if it has an amplitude at least twice that of the other secondary peak associated with the primary peak.

15. The method of claim 14, further including selecting the earlier occurring of said secondary peaks.

16. The method of claim 1, further including comparing secondary peaks of said first waveform trace with secondary peaks of said second waveform trace to determine reproducibility therebetween, and where a secondary peak of one said waveform trace is reproducible with a pair of secondary peaks of the other waveform trace, selecting the earlier occurring secondary peak of the pair as being reproducible.

17. The method of claim 1, wherein for analyzing brain stem signal waveforms for a prominent wave, said reproducibility comprises a window of about 0.25 milliseconds.

18. The method of claim 1, further including defining a prominent wave if there exists a single pair of reproducible secondary peaks associated with said first and second waveform traces.

19. The method of claim 1, further including comparing amplitudes of reproducible secondary peaks of each waveform trace, and selecting a reproducible secondary peak which has an amplitude more than twice the amplitude of a next highest reproducible secondary peak amplitude.

20. The method of claim 19, further including selecting as said secondary peak of said multiple which is more aligned in time with the secondary peak of the other waveform trace.

21. The method of claim 20, further including selecting the earlier occurring secondary peak of said multiple if both secondary peaks of said multiple are substantially tied as to time alignment with the secondary peak of the other waveform trace.

22. The method of claim 21, further including comparing characteristics of said reproducible secondary peaks, calculating a ratio of preceding valley to secondary peak and secondary peak to succeeding valley of each reproducible secondary peak, and defining said prominent wave as the reproducible pair having a smallest sum of said ratios.

23. The method of claim 1, further including carrying out said steps a first time with data filtered according to a predefined bandpass, and if no prominent wave is found, filtering said data with a narrower bandpass and again carrying out said steps.

24. The method of claim 1, further including identifying Wave V for a brain stem signal waveform, said analyzing being carried out on data having a latency between about 5.2-10.5 milliseconds.

25. The method of claim 1, wherein for identifying Wave I of a brain stem signal waveform, said steps are carried out with a latency having a lower limit of about 1.4 milliseconds, and an upper limit of about 3.0 milliseconds if a Wave V is not found, and an upper limit of the Wave V upper latency limit minus about 3.5 milliseconds.

26. The method of claim 1, further including visually displaying said signal waveform, and locating a cursor adjacent a point on said waveform which is analyzed as being said prominent waveform.

27. The method of claim 1, further including identifying said prominent wave which corresponds to said selected reproducible secondary peaks, visually displaying said first and second waveform traces, and visually displaying a calculated latency of a displayed prominent wave of each said first and second waveform traces.

28. A computerized signal waveform processing system, comprising:

a filter having a plurality of selectable bandpasses;

a program controlled processor for storing and processing digital signals corresponding to signal waveform;

an output device controlled by said processor for displaying results of processed digital signals;

a program having instructions carried out by said processor for analyzing said digital signals, said program being operative to;

filter said digital signals according to selected bandpasses;

analyze said filtered digital signals to define primary peaks of said waveform;

analyze digital data representative of said primary peaks to determine secondary peaks;

analyze time and amplitude characteristics of said secondary peaks to define a location on said primary peak for identification thereof; and

provide an output on said output device for indicating a location on said signal waveform which identifies a prominent wave of said waveform.

29. The computerized equipment of claim 28, further including a software routine for filtering said digital signals according to a predefined bandpass characteristic.

30. The computerized equipment of claim 28, wherein said output identifying said prominent wave comprises a cursor.

31. The computerized equipment of claim 28, wherein said output identifying said prominent wave comprises a numerical readout defining a latency point of said prominent wave.

32. A method for analyzing brain stem waveform traces using a computerized signal waveform processing system so as to identify a prominent wave, comprising the steps of:

storing and processing digital signals using a program controlled processor;

displaying the processed digital signals;

analyzing said digital signals using a program having instructions carried out by said processor, said analyzation including;

analyzing digital signals of related first and second waveform traces to define primary peaks which satisfy a predetermined criteria, each said primary peak having associated preceding and succeeding troughs;

if a primary peak is not determined by initial analyzation of said digital signals, filtering said digital signals with a predefined bandpass and again analyzing the filtered digital signals until one said primary peak is found in each said waveform trace;

analyzing said primary peaks to determine secondary peaks;

analyzing said secondary peaks to characterize said secondary peaks according to predefined criteria;

determining reproducibility of said secondary peaks between said first and second waveform traces; and

identifying said prominent wave by employing reproducibility criteria for the secondary peaks of said first and second waveform traces.

33. The method of claim 32, further including analyzing said digital signals between a latency period of about 5.2-10.5 milliseconds for identifying a Wave V of a brain stem signal waveform.

34. The method of claim 32, further including filtering said digital data when a primary peak is not found, and using an ordered filter sequence of bandpass frequencies of:

(a) Bandpass 1-8000 Hz

(b) Bandpass 1-3000 Hz

(c) Bandpass 1-2000 Hz

(d) Bandpass 1-1500 Hz

(e) Bandpass 1-1000 Hz

wherein the sequence of bandpasses are used until a primary peak is found.

35. The method of claim 32, further including determining said reproducibility by comparing, time-wise, secondary peaks of the first waveform trace with secondary peaks of the second waveform trace, and selecting a secondary peak on the first waveform trace that is within about 0.25 milliseconds of a secondary peak on the second waveform trace as a reproducible pair.

36. The method of claim 35, wherein for multiple reproducible pairs, further including choosing the reproducible pair that occurs earlier in time as said prominent wave when the reproducibility criteria analysis employed to identify said prominent wave determines that multiple reproducible pairs qualify as a prominent wave.