Method and computer program product for assessing neurological conditions and treatments using evoked response potentials

• US 6,463,321 B2
• Filed: 02/02/2001
• Issued: 10/08/2002
• Est. Priority Date: 08/10/1999
• Status: Expired due to Fees

• Current Assignee: Regents of the University of California (University of California)
• Original Assignee: Thuris Corp. (Protein Polymer Technologies, Inc.)
• Inventors: Granger, Richard
• Primary Examiner(s): Shaver, Kevin
• Assistant Examiner(s): Natnithithadha, Navin
• Application Number:
  US09/773,732
• Publication Number:
  US 20010020137A1
• Time in Patent Office: 613 Days
• Field of Search: 600/544, 600/545, 600/546, 128/920, 128/923
• US Class Current: 600/544
• CPC Class Codes: A61B 5/377 using evoked responses
• CPC Class Codes: A61B 5/389 Electromyography [EMG]
• CPC Class Codes: A61B 5/7246 using correlation, e.g. template matching or determination of similarity
• CPC Class Codes: G16H 20/30 relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
• CPC Class Codes: G16H 40/63 for local operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients
• CPC Class Codes: Y10S 128/92 Computer assisted medical diagnostics

Abstract

A method is provided for diagnosing the presence of a neurological disorder or otherwise assessing the neurological condition of a patient. The method also allows the assessment of a treatment regimen used by a patient. The method includes the collection and analysis of ERP data. The method of the invention begins by conducting a plurality of ERP trials on a patient. In an embodiment of the invention, the data from the ERP trials is then characterized to produce a characterizing ERP signal vector for the patient. Projections based on the characterizing ERP signal vector are then generated. The projections are compared to information derived from the ERP data of patients having known neurological conditions. To perform diagnosis, the projections are compared to standards, such as one or more characterizing ERP signal vectors from known healthy patients, and one or more characterizing ERP signal vectors from patients known to have the disorder. The probable presence or absence of the neurological disorder is decided by a weighted vote of the projections, where the weighting is a function of how closely each projection compares to the respective standards. Projections can also be used to perform other types of neurological assessment, such as tracking a patient'"'"'s response to a treatment regimen, assessing the treatability of a patient with respect to a particular regimen, determining the effects of a particular regimen, or categorizing a patient in order to create a homogenous group for a clinical trial.

Claims

1. A method of facilitating the diagnosis of a neurological disorder, comprising the steps of:

(a) performing a plurality of evoked response potential (ERP) trials on a patient under assessment;

(b) generating a plurality of projections of ERP data, wherein the ERP data is derived from the plurality of ERP trials performed on the patient under assessment; and

(c) conducting a weighted vote process that is based on the projections, to aid in determining the presence of the neurological disorder.

2. The method of claim 1, wherein said step (a) comprises the step of repeating an ERP trial approximately once every one to two seconds, for approximately two minutes.

3. The method of claim 1, further comprising the step of:

(d) characterizing ERP signals obtained from the plurality of trials performed on the patient under assessment, to produce a characterizing ERP signal vector for the patient under assessment;

wherein said step (d) is performed after said step (a) and before said step (b), and

wherein said step (b) comprises generating a plurality of projections of the characterizing ERP signal vector of the patient under assessment.

4. The method of claim 3, wherein said step (d) comprises the steps of:

(i) for each ERP trial, concatenating the signals received by electrodes of an ERP trial apparatus in a predetermined order to form a concatenated signal;

(ii) for each ERP trial, sampling the concatenated signal at a predetermined sampling rate, to form a vector of amplitude values for each ERP trial; and

(iii) averaging corresponding amplitude values taken from each vector, to form a characterizing ERP signal vector of the patient under assessment.

5. The method of claim 4, wherein said step ii) comprises sampling each concatenated signal approximately once every 10 milliseconds.

6. The method of claim 3, wherein said step (b) comprises randomly deleting components of the characterizing ERP signal vector of the patient under assessment to create the projections.

7. The method of claim 3, wherein said step (b) comprises deleting components of the characterizing ERP signal vector of the patient under assessment according to a deterministic algorithm to create the projections.

8. The method of claim 1, wherein said step (c) comprises the step of:

(i) comparing each projection with characterizing ERP signal vectors from other patients, to produce comparison results for each projection.

9. The method of claim 8, wherein said step (i) comprises comparing each projection with at least one characterizing ERP signal vector from at least one patient, respectively, known not to be afflicted with the neurological disorder.

10. The method of claim 8, wherein said step (i) comprises comparing each projection with at least one characterizing ERP signal vector from at least one patient, respectively, known to be afflicted with the neurological disorder.

11. The method of claim 8, wherein said step (i) comprises the steps of:

(A) deriving a healthy weight value for each projection and each characterizing ERP signal vector from a healthy patient, wherein the healthy weight value is determined by a degree of correlation between the projection and the characterizing ERP signal vector from the healthy patient, such that a greater degree of correlation between the projection and the characterizing ERP signal vector from the healthy patient corresponds to a larger healthy weight value;

(B) deriving a disorder weight value for each projection and each characterizing ERP signal vector from a patient known to have the neurological disorder, wherein the disorder weight value is determined by a degree of correlation between the projection and the characterizing ERP signal vector from the patient known to have the neurological disorder, such that a greater degree of correlation between the projection and the characterizing ERP signal vector from the patient known to have the neurological disorder corresponds to a larger disorder weight value;

(C) for each characterizing ERP signal vector from a healthy patient, summing the healthy weight values for all projections, to produce a healthy sum corresponding to the associated healthy patient;

(D) for each characterizing ERP signal vector from a patient known to have the neurological disorder, summing the disorder weight values for all projections, to create a disorder sum corresponding to the associated patient known to have the neurological disorder;

(E) comparing the healthy sums and the disorder sums; and

(F) determining whether any of the disorder sums is greater than any of the healthy sums.

12. A computer program product comprising a computer usable medium having computer readable program code means embodied in the medium for causing an application program to execute on a computer that facilitates the diagnosis of a neurological disorder, the computer readable program code means comprising:

(a) first computer readable program code means for causing the computer to characterize evoked response potential (ERP) signals obtained from a plurality of ERP trials conducted on a patient under assessment, to produce a characterizing ERP signal vector of the patient under assessment;

(b) second computer readable program code means for causing the computer to generate a plurality of projections of the characterizing ERP signal vector of the patient under assessment;

(c) third computer readable program code means for causing the computer to compare each projection with characterizing ERP signal vectors from other patients; and

(d) fourth computer readable program code means for causing the computer to conduct a weighted vote process that is based on comparison results of each projection, to aid in determining the probable presence of the neurological disorder.

13. A method of facilitating determination of treatability, with respect to a first treatment regimen, of a patient under assessment, the method comprising the steps of:

(a) performing a plurality of evoked response potential (ERP) trials on the patient under assessment;

(b) generating a plurality of projections of ERP data, wherein the ERP data is derived from the plurality of ERP trials performed on the patient under assessment; and

(c) comparing the plurality of projections to data collected from at least one other patient, to aid in assessing the likely responsiveness of the patient under assessment to the first treatment regimen.

14. The method of claim 13, wherein said step (a) comprises the step of repeating an ERP trial approximately once every one to two seconds, for approximately 2 minutes.

15. The method of claim 13, further comprising the step of:

(d) characterizing ERP signals obtained from the plurality of trials performed on the patient under assessment, to produce a characterizing ERP signal vector of the patient under assessment;

wherein said step (d) is performed after said step (a) and before said step (b), and wherein said step (b) comprises generating a plurality of projections of the characterizing ERP signal vector of the patient under assessment.

16. The method of claim 15 wherein said step (d) comprises the steps of:

(i) for each ERP trial performed on the patient under assessment, concatenating the signal received by electrodes of an ERP trial apparatus in a predetermined order, to form a concatenated signal;

(ii) for each ERP trial performed on the patient under assessment, sampling the concatenated signal at a predetermined sampling rate, to form a vector of amplitude values for each ERP trial; and

(iii) averaging corresponding amplitude values taken from each vector, to form a characterizing ERP signal vector of the patient under assessment.

17. The method of claim 16, wherein said step (ii) comprises sampling each concatenated signal approximately once every ten milliseconds.

18. The method of claim 15, wherein said step (b) comprises randomly deleting components of the characterizing ERP signal vector of the patient under assessment to create the projections.

19. The method of claim 15, wherein said step (b) comprises deleting components of the characterizing ERP signal vector of the patient under assessment according to a deterministic algorithm to create the projections.

20. The method of claim 13, wherein said step (c) comprises the step of:

(i) comparing each projection with at least one characterizing ERP signal vector from at least one other patient, respectively, known to be treatable by the first treatment regimen.

21. The method of claim 20, wherein said step (i) comprises the steps of:

(A) for each projection and each of at least one characterizing ERP signal vector from each of at least one other patient, respectively, known to be treatable by the first treatment regimen, deriving a first weight value that is determined by a degree of correlation between the projection and the characterizing ERP signal vector from the patient known to be treatable by the first treatment regimen; and

(B) for each characterizing ERP signal vector from a patient known to be treatable by the first treatment regimen, summing the first weight values for all projections to produce a first sum to aid in deciding on the probable responsiveness of the patient under assessment to the first treatment regimen, based on the first sums.

22. The method of claim 21, wherein said step (i) further comprises the steps of:

(C) for each projection and each of at least one characterizing ERP signal vector from at least one other patient, respectively, known to be treatable by a second treatment regimen, deriving a second weight value for each projection, wherein the second weight value is determined by a degree of correlation between the projection and the characterizing ERP signal vector from the patient known to be treatable by the second treatment regimen;

(D) for each characterizing ERP signal vector from a patient known to be treatable by the second treatment regimen, summing the second weight values for all projections, to produce a second sum; and

(E) comparing the first sums and the second sums to aid in deciding on the likely responsiveness of the patient under assessment to the first treatment regimen, relative to the likely responsiveness to the second treatment regimen, based on the first and second sums.

23. A computer program product comprising a computer usable medium having computer readable program code means embodied in the medium for causing an application program to execute on a computer that facilitates the assessment of the likely responsiveness of a patient under assessment to a treatment regimen, the computer readable program code means comprising:

(a) first computer readable program code means for causing the computer to characterize evoked response potential (ERP) signals obtained from a plurality of ERP trials, to produce a characterizing ERP signal vector of the patient under assessment;

(b) second computer readable program code means for causing the computer to generate a plurality of projections of the characterizing ERP signal vector of the patient under assessment; and

(c) third computer readable program code means for causing the computer to compare each projection with characterizing ERP signal vectors from other patients to aid in determining the probable treatability of the patient having a neurological disorder with respect to the treatment regimen, where the determination is based on comparison results of each projection.

24. A method of monitoring the response, to a treatment regimen, of a patient being treated for a neurological disorder, the method comprising the steps of:

(a) performing a plurality of evoked response potential (ERP) trials on the patient being treated;

(b) generating a plurality of projections of ERP data, wherein the ERP data is derived from the plurality of ERP trials performed on the patient being treated; and

(c) comparing the plurality of projections to data collected from at least one other patient, to aid in assessing the current condition of the patient being treated, as indicated by the projections, after beginning the treatment regimen.

25. The method of claim 24 wherein said step (a) comprises the step of repeating an ERP trial approximately once every one to two seconds, for approximately two minutes.

26. The method of claim 24, further comprising the step of:

(d) characterizing ERP signals obtained from the plurality of trials performed on the patient being treated, to produce a characterizing ERP signal vector of the patient being treated;

wherein said step (d) is performed after said step (a) and before said step (b), and wherein said step (b) comprises generating a plurality of projections of the characterizing ERP signal vector of the patient being treated.

27. The method of claim 26, wherein said step (d) comprises the steps of:

(i) for each ERP trial, concatenating the signals received by electrodes of an ERP trial apparatus in a predetermined order to form a concatenated signal;

(ii) for each ERP trial, sampling the concatenating signal at a predetermined sampling rate, to form a vector of amplitude values for each ERP trial; and

(iii) averaging corresponding amplitude values taken from each vector, to form a characterizing ERP signal vector of the patient being treated.

28. The method of claim 27 wherein said step (ii) comprises sampling each concatenated signal approximately once every ten milliseconds.

29. The method of claim 26, wherein said step (b) comprises randomly deleting components of the characterizing ERP signal vector of the patient being treated, to create the projections.

30. The method of claim 26, wherein said step (b) comprises deleting components of the characterizing ERP signal vector of the patient being treated, wherein deletions are performed according to a deterministic algorithm to create the projections.

31. The method of claim 24, wherein said step (c) comprises the step of:

(i) comparing each projection with at least one characterizing ERP signal vector from at least one other patient, respectively, wherein each such patient has a known neurological condition.

32. The method of claim 31, wherein said step (i) comprises the steps of:

(A) for each of at least one characterizing ERP signal vector from at least one patient, respectively, and for each projection, deriving a progress weight value, wherein the progress weight value is determined by a degree of correlation between the projection and the characterizing ERP signal vector from a patient having a known neurological condition; and

(B) for each of at least one characterizing ERP signal vector from at least one patient, respectively, summing the progress weight values for all projections, to produce a progress sum to aid in deciding on the probable current condition of the patient being treated, based on the progress sums.

33. The method of claim 31, wherein step (i) comprises the step of comparing each projection with at least one characterizing ERP signal vector from at least one other patient, respectively, wherein at least one such patient comprises a patient who is neurologically healthy.

34. The method of claim 31, wherein step (i) comprises the step of comparing each projection with at least one characterizing ERP signal vector from at least one other patient, respectively, wherein at least one such patient comprises a patient who is afflicted with the neurological disorder of the patient being treated.

35. A computer program product comprising a computer usable medium having computer readable code means embodied in the medium for causing an application program to execute on a computer that facilitates the assessment of a current condition of a patient being treated for a neurological disorder after beginning a treatment regimen, the computer readable code means comprising:

(a) first computer readable program code means for causing the computer to characterize evoked response potential (ERP) signals obtained from a plurality of ERP trials performed on the patient being treated, to produce a characterizing ERP signal vector of the patient being treated;

(b) second computer readable program code means for causing the computer to generate a plurality of projections of the characterizing ERP signal vector of the patient being treated; and

(c) third computer readable program code means for causing the computer to compare each projection with characterizing ERP signal vectors from other patients to aid in determining the probable current condition of the patient being treated, where the determination is based on comparison results of each projection.

36. A method of assessing a patient with respect to the nature and extent of side effects experienced in response to a treatment regimen, wherein the patient is being treated for a neurological disorder, the method comprising the steps of:

(a) performing a plurality of evoked response potential (ERP) trials on the patient under assessment;

(b) generating a plurality of projections of ERP data, wherein the ERP data is derived from the plurality of ERP trials performed on the patient under assessment; and

(c) comparing the plurality of projections to data collected from at least one other patient, to aid in assessing the nature and extent of the side affects as indicated by the projections.

37. The method of claim 36 wherein said step (a) comprises the step of repeating an ERP trial approximately once every one to two seconds, for approximately two minutes.

38. The method of claim 36, further comprising the step of:

(d) characterizing ERP signals obtained from the plurality of trials performed on the patient under assessment, to produce a characterizing ERP signal vector of the patient under assessment;

wherein said step (d) is performed after said step (a) and before said step (b), and wherein said step (b) comprises generating a plurality of projections of the characterizing ERP signal vector of the patient under assessment.

39. The method of claim 38, wherein said step (d) comprises the steps of:

(i) for each ERP trial, concatenating the signals received by electrodes of an ERP trial apparatus in a predetermined order to form a concatenated signal;

(ii) for each ERP trial, sampling the concatenated signal at a predetermined sampling rate, to form a vector of amplitude values for each ERP trial; and

(iii) averaging corresponding amplitude values taken from each vector, to form a characterizing ERP signal vector of the patient under assessment.

40. The method of claim 39, wherein said step (ii) comprises sampling each concatenated signal approximately once every ten milliseconds.

41. The method of claim 38, wherein said step (b) comprises randomly deleting components of the characterizing ERP signal vector of the patient under assessment to create the projections.

42. The method of claim 38, wherein said step (b) comprises deleting components of the characterizing ERP signal vector of the patient under assessment according to a deterministic algorithm to create the projections.

43. The method of claim 36, wherein step (c) comprises the step of:

(i) comparing each projection with at least one characterizing ERP signal vector from at least one other patient, respectively, wherein the at least one other patient has at least one known neurological condition, respectively.

44. The method of claim 43, wherein step (i) comprises the steps of:

(A) for each characterizing ERP signal vector from a patient having a known neurological condition and for each projection, deriving a side effect weight value, wherein each side effect weight value is determined by a degree of correlation between the projection and the characterizing ERP signal vector of the patient having a known neurological condition;

(B) for each characterizing ERP signal vector from a patient having a known neurological condition, summing the side affect weight values for all projections to produce a side effect sum to aid in deciding on the probable nature and extent of the side effects, based on the side effect sums.

45. A computer program product comprising a computer usable medium having computer readable program code means embodied in the medium for causing an application program to execute on a computer that facilitates the assessment of the nature and extent of side affects resulting from a treatment regimen, the computer readable program code means comprising:

(a) first computer readable program code means for causing the computer to characterize evoked response potential (ERP) signals obtained from a plurality of ERP trials performed on a patient under assessment, to produce a characterizing ERP signal vector of the patient under assessment;

(b) second computer readable program code means for causing the computer to generate a plurality of projections of the characterizing ERP signal vector of the patient under assessment; and

(c) third computer readable program code means for causing the computer to compare each projection with characterizing ERP signal vectors from other patients to aid in determining the probable nature and extent of side affects resulting from the treatment regimen, wherein the determination is based on comparison results of each projection.

46. A method of assessing the results of a treatment regimen for a neurological disorder, comprising the steps of:

(a) performing a plurality of evoked response potential (ERP) trials on a test patient afflicted with the neurological disorder and treated with the test regimen;

(b) generating a plurality of projections of ERP data, wherein the ERP data is derived from the plurality of ERP trials performed on the test patient; and

(c) comparing the plurality of projections to data collected from at least one other patient, to aid in determining the neurological condition of the test patient as indicated by the projections.

47. The method of claim 46 wherein said step (a) comprises the step of repeating an ERP trial approximately once every one to two seconds, for approximately two minutes.

48. The method of claim 46, further comprising the step of:

(d) characterizing ERP signals obtained from the plurality of trials performed on the test patient, to produce a characterizing ERP signal vector of the test patient;

wherein said step (d) is performed after said step (a) and before said step (b), and wherein said step (b) comprises generating a plurality of projections of the characterizing ERP signal vector of the test patient.

49. The method of claim 48, wherein said step (d) comprises the steps of:

(i) for each ERP trial, concatenating the signals received by electrodes of an ERP trial apparatus in a predetermined order to form a concatenated signal;

(ii) for each ERP trial, sampling the concatenated signal at a predetermined sampling rate, to form a vector of amplitude values for each ERP trial; and

(iii) averaging corresponding amplitude values taken from each vector, to form a characterizing ERP signal vector of the test patient.

50. The method of claim 49, wherein said step (ii) comprises sampling each concatenated signal approximately once every ten milliseconds.

51. The method of claim 48, wherein said step (b) comprises randomly deleting components of the characterizing ERP signal vector of the test patient to create the projections.

52. The method of claim 48, wherein said step (b) comprises deleting components of the characterizing ERP signal vector of the test patient according to a deterministic algorithm to create the projections.

53. The method of claim 46, wherein step (c) comprises the step of:

(i) comparing each projection with at least one characterizing ERP signal vector from at least one other patient, respectively, wherein the at least one other patient has at least one known neurological condition, respectively.

54. The method of claim 53, wherein step (i) comprises the steps of:

(A) for each characterizing ERP signal vector from a patient having a known neurological condition, and for each projection, deriving a comparison weight value, wherein each comparison weight value is determined by a degree of correlation between the projection and the characterizing ERP signal vector of the patient having a known neurological condition; and

(B) for each characterizing ERP signal vector from a patient having a known neurological condition, summing the comparison weight values for all projections to produce a comparison sum to aid in characterizing the effects of the treatment regimen, based on the comparison sums.

55. A computer program product comprising a computer usable medium having computer readable program code means embodied in the medium for causing an application program to execute on a computer that facilitates the assessment of the results of a treatment regimen for a neurological disorder, the computer readable program code means comprising:

(a) first computer readable program code means for causing the computer to characterize evoked response potential (ERP) signals obtained from a plurality of ERP trials performed on a test patient to produce a characterizing ERP signal vector of the test patient;

(b) second computer readable program code means for causing the computer to generate a plurality of projections of the characterizing ERP signal vector of the test patient; and

(c) third computer readable program code means for causing the computer to compare each projection with characterizing ERP signal vectors from other patients to aid in determining the probable neurological condition of the test patient resulting from the treatment regimen, wherein the determination is based on comparison results of each projection.

56. A method of determining an enrollment of subjects in a study of a neurological treatment, comprising the steps of:

(a) performing a plurality of evoked response potential (ERP) trials on a candidate;

(b) plotting the candidate in an information space on the basis of data collected during the ERP trials;

(c) identifying the candidate as an outlying candidate if the plot of the candidate is sufficiently distant from a plot representing at least one other candidate; and

(d) identifying the outlying candidate as a candidate to be excluded from the study.

57. The method of claim 56, wherein said step (a) comprises the step of repeating an ERP trail approximately once every one to two seconds, for approximately two minutes.

58. The method of claim 56, further comprising the step of:

(e) characterizing ERP signals obtained from the plurality of ERP trails, to produce a characterizing ERP signal vector, wherein said step_(e) is performed after said step_(a) and before said step_(b), and wherein said step (b) comprises plotting the candidate in the information space on the basis of the characterizing ERP signal vector.

59. The method of claim 58, wherein said step (e) comprises the steps of:

(i) for each ERP trial, concatenating the signals received by electrodes of an ERP trial apparatus in a predetermined order to form a concatenated signal;

(ii) for each ERP trial, sampling the concatenated signal at a predetermined sampling rate, to form a vector of amplitude values for each ERP trial; and

(iii) averaging corresponding amplitude values taken from each vector, to form a characterizing ERP signal vector of the candidate.

60. The method of claim 59, wherein said step (ii) comprises sampling each concatenated signal approximately once every ten milliseconds.

61. The method of claim 56, wherein the plot representing the at least one other candidate is a median point for the at least one other candidate.

62. The method of claim 57, wherein the plot representing the at least one other candidate is a mean point for the at least one other candidate.

63. A computer program product comprising a computer-usable medium having computer-readable program code means embodied in the medium for causing an application program to execute on a computer that facilitates the determination of enrollment of subjects in a study of a neurological treatment, the computer-readable program code means comprising:

(a) first computer-readable code means for causing the computer to characterize evoked response potential (ERP) signals obtained from a plurality of ERP trials performed on a candidate to produce a characterizing ERP signal vector of the candidate;

(b) second computer-readable code means for causing the computer to plot the candidate in an information space on the basis of the characterizing ERP signal vector;

(c) third computer-readable program code means for causing the computer to identify the candidate as an outlying candidate if the plot of the candidate is sufficiently distant from a plot representing at least one other candidate.