METHOD, SYSTEM, AND COMPUTER-ACCESSIBLE MEDIUM FOR CLASSIFICATION OF AT LEAST ONE ICTAL STATE

• US 20110218950A1
• Filed: 06/02/2009
• Published: 09/08/2011
• Est. Priority Date: 06/02/2008
• Status: Active Grant

• Current Assignee: New York University
• Original Assignee: New York University
• Inventors: Kuzniecky, Ruben, Lecun, Yann, Mirowski, Piotr W., Madhavan, Deepak
• Granted Patent: US 9,443,141 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: A61B 5/245 specially adapted for magnetoencephalographic [MEG] signals
• CPC Class Codes: A61B 5/318 Heart-related electrical modalities, e.g. electrocardiography [ECG]
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/372 Analysis of electroencephalograms
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: G06F 2218/12 Classification; Matching
• CPC Class Codes: G06N 20/10 using kernel methods, e.g. support vector machines [SVM]
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06N 3/044 Recurrent networks, e.g. Hopfield networks
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 3/084 Backpropagation, e.g. using gradient descent
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

An exemplary methodology, procedure, system, method and computer-accessible medium can be provided for receiving physiological data for the subject, extracting one or more patterns of features from the physiological data, and classifying the at least one state of the subject using a spatial structure and a temporal structure of the one or more patterns of features, wherein at least one of the at least one state is an ictal state.

Claims

1. A method for classifying at least one state of a subject, comprising:

receiving physiological data for the subject;

extracting one or more patterns of features from the physiological data; and

using a computing arrangement, classifying the at least one state of the subject using a spatial structure and a temporal structure of the one or more patterns of features, wherein at least one of the at least one state is an ictal state.

14. The method of claim 1, wherein the one or more patterns of features are spatially-varying and time-varying.

18. The method of claim 1, wherein the temporal structure of the one or more patterns of features comprises a local time structure.

19. The method of claim 1, wherein the features comprise bivariate features.

23. The method of claim 1, wherein the classifying procedure is performed using at least one of a logistic regression or a support vector machine.

32. The method of claim 1, further comprising:

electronically storing the received physiological data;

receiving further physiological data for the subject;

extracting one or more further patterns of features from the further physiological data of the subject; and

using the computing arrangement, classifying the at least one state of the subject based on the patterns of features from the received physiological data and the further patterns of features based on the further physiological data.

53. The system of claim 1, wherein the physiological data is at least one of EEG data, multi-channel EEG data, fMR1 data, MEG data, EKG data, pulse data, respiration data, temperature data, eye movement data or blood chemistry data.

54. The system of claim 53, wherein the one or more patterns of features are spatially-varying and time-varying.

55. The system of claim 54, wherein the one or more patterns of features are frequency-varying.

2-13. -13. (canceled)

15-17. -17. (canceled)

20-22. -22. (canceled)

24-31. -31. (canceled)

33. A computer-accessible medium for classifying at least one state of a subject, the computer-accessible medium including instructions thereon, wherein, when a computing arrangement executes the instructions, the computing arrangement is configured to perform procedures comprising:

receiving physiological data for the subject;

extracting one or more patterns of features from the physiological data of the subject; and

using a computing arrangement for classifying the at least one state of the subject using at least one of a spatial structure and a temporal structure of the one or more patterns of features, wherein at least one of the at least one state is an ictal state.

34. The computer-accessible medium of claim 33, wherein the one or more patterns of features is represented by two-dimensional data.

35. The computer-accessible medium of claim 33, wherein the one or more patterns of features is represented by three-dimensional data.

36. The method of claim 33, wherein the computing arrangement classifies the at least one ictal state of the subject using convolutional networks.

37. The computer-accessible medium of claim 33, wherein the at least one ictal state comprises at least one of an ictal classification, a peri-ictal classification, a pre-ictal classification or an interictal classification.

38. The computer-accessible medium of claim 33, wherein the features comprise bivariate features.

39. The computer-accessible medium of claim 33, wherein the processing arrangement is further configured for maintaining a memory of previously received physiological data after classifying the ictal state.

40. A system for classifying at least one state of a subject, comprising:

an input module, which, when executed by a computing arrangement, causes the computing arrangement to obtain one or more first extracted patterns of features from physiological data of a subject at a particular time interval; and

a classifier module, which, when executed by the computing arrangement, causes the computing arrangement to classify at least one state of the subject using at least one of a spatial and temporal structure of the one or more extracted patterns of features, wherein at least one of the at least one state is an ictal state.

43. The system of claim 40, wherein the classifier module is a trained classifier module.

49. The system of claim 40, wherein the one or more patterns of features is represented by three-dimensional data.

50. The system of claim 49, wherein the three-dimensional data is data of time versus a pair of channels versus frequency.

51. The system of claim 40, wherein the classifier module uses convolutional networks for classifying the at least one ictal state of the subject.

52. The system of claim 40, wherein the at least one ictal state comprises at least one of an ictal classification, a peri-ictal classification, a pre-ictal classification or an interictal classification.

57. The system of claim 40, wherein the classifier discriminates between pre-ictal and inter-ictal states.

58. The system of claim 40, wherein the temporal structure of the one or more patterns of features comprises a local time structure.

59. The system of claim 40, wherein the features comprise bivariate features.

60. The system of claim 59, wherein the bivariate features are based on a measure of a synchronization among one or more subsets of the physiological data.

61. The system of claim 60, wherein the measure of the synchronization is based on at least one of a cross-correlation, a nonlinear interdependence, a difference of Lyapunov exponents or a phase-locking synchrony.

62. The system of claim 61, wherein the synchrony comprises one of a phase-locking statistic, an entropy of a phase difference and distribution, or a coherence.

63. The system of claim 40, wherein the classifier uses at least one of a logistic regression or a support vector machine.

65. The system of claim 40, wherein the classifier is regularized using a lasso regularization or an L1-norm regularization.

66. The system of claim 40, wherein the classifier is regularized using a ridge regularization or an L2-norm regularization.

67. The system of claim 40, wherein the classifier is trained using a stochastic gradient descent.

68. The system of claim 40, wherein the classifier is trained by determining finite impulse response filters of at least a portion of the features.

69. The system of claim 68, wherein the features are consecutive features.

70. The system of claim 40, further comprising performing a sensitivity analysis on the patterns of features before classifying the at least one state.

71. The system of claim 70, wherein the sensitivity analysis is used to evaluate an importance of a particular feature according to a spatial location, time or a frequency location in the pattern of features thereof.

72. The system of claim 66, wherein the classifier is a support vector machine and the patterns of features that are in a training dataset and that serve as support vectors for the classifier are used to evaluate an electrophysiological recording or the development of ictal states over time.

74. The system of claim 60, wherein the measure of the synchronization is based on wavelet analysis-based synchrony values grouped in one or more frequency bands.

41-42. -42. (canceled)

44-48. -48. (canceled)

56. (canceled)

64. (canceled)