Detection of hypokinetic and hyperkinetic states
First Claim
1. An automated method of determining a bradykinetic state of a person, which is a state characterized by slowness of movement relative to a normally kinetic person and in which movements involving peaks of acceleration have strong low frequency components, the method comprising:
- obtaining a time series of accelerometer data in real-time from an accelerometer worn on an extremity of the person, the time series of accelerometer data being representative of movement of the person over an extended period of time, wherein the time series of accelerometer data is obtained while the accelerometer is oriented to be sensitive to at least one of pronation and supination movements of the person and is attached to at least a limb of the person;
filtering, by a band pass filter, the accelerometer data to extract filtered data for a band of interest, wherein the band of interest has a lower end cut off frequency selected to remove DC;
extracting, by a computer processor, a plurality of bins of acceleration data from the time series of the filtered accelerometer data;
searching, by the computer processor, the data in each of the plurality of bins to identify a time window of peak acceleration, the time window within a given bin comprising at least two data samples that capture accelerometer data for a single movement having a peak acceleration relative to other movements observed within the plurality of bins; and
processing, by the computer processor, data in the time windows of peak acceleration to determine;
spectral components for each of the single movements having peak acceleration; and
an objective measure of the extent to which the single movements having peak acceleration identified within the plurality of bins of acceleration data are slower than expected from a normally kinetic person and indicative of bradykinesia based upon a comparative statistical analysis of the determined sets of spectral components of the single movements of peak acceleration that occur over the extended period of time.
6 Assignments
0 Petitions
Accused Products
Abstract
The present invention relates to an automated method of determining a kinetic state of a person. The method obtains accelerometer data from an accelerometer worn on an extremity of the person and processes the accelerometer data to determine a measure for the kinetic state. The present invention further relates to a device for determining a kinetic state of a person. The device comprises a processor configured to process data obtained from an accelerometer worn on an extremity of the person and to determine from the data a measure for the kinetic state. In the method and system the kinetic state is at least one of bradykinesia, dyskinesia, and hyperkinesia.
-
Citations
17 Claims
-
1. An automated method of determining a bradykinetic state of a person, which is a state characterized by slowness of movement relative to a normally kinetic person and in which movements involving peaks of acceleration have strong low frequency components, the method comprising:
-
obtaining a time series of accelerometer data in real-time from an accelerometer worn on an extremity of the person, the time series of accelerometer data being representative of movement of the person over an extended period of time, wherein the time series of accelerometer data is obtained while the accelerometer is oriented to be sensitive to at least one of pronation and supination movements of the person and is attached to at least a limb of the person; filtering, by a band pass filter, the accelerometer data to extract filtered data for a band of interest, wherein the band of interest has a lower end cut off frequency selected to remove DC; extracting, by a computer processor, a plurality of bins of acceleration data from the time series of the filtered accelerometer data; searching, by the computer processor, the data in each of the plurality of bins to identify a time window of peak acceleration, the time window within a given bin comprising at least two data samples that capture accelerometer data for a single movement having a peak acceleration relative to other movements observed within the plurality of bins; and processing, by the computer processor, data in the time windows of peak acceleration to determine; spectral components for each of the single movements having peak acceleration; and an objective measure of the extent to which the single movements having peak acceleration identified within the plurality of bins of acceleration data are slower than expected from a normally kinetic person and indicative of bradykinesia based upon a comparative statistical analysis of the determined sets of spectral components of the single movements of peak acceleration that occur over the extended period of time. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
16. A wearable device to assist in determining a bradykinetic state of a person, which is a state characterized by slowness of movement relative to a normally kinetic person and in which movements involving peaks of acceleration have strong low frequency components, the device comprising:
-
an accelerometer arranged to be worn on at least a limb of the person, wherein the accelerometer is oriented to be sensitive to at least one of pronation and supination movements of the person; a band pass filter for band pass filtering a time series of accelerometer data obtained in real-time from the accelerometer worn on at least the limb of the person to extract filtered data for a band of interest, the time series of accelerometer data being representative of movement of the person over an extended period of time, wherein the band of interest has a lower end cut off frequency selected to remove DC; and a processor configured to perform operations comprising; extracting a plurality of bins of filtered accelerometer data from the time series of the accelerometer data, searching the data in each of the plurality of bins to identify a time window of peak acceleration, the time window within a given bin comprising at least two data samples that capture filtered accelerometer data for a single movement having a peak acceleration relative to other movements observed within the given bin, and processing the data in the time windows of peak acceleration to determine; spectral components for each of the single movements having peak acceleration; and an objective measure of the extent to which the single movements having peak acceleration identified within the plurality of bins of filtered accelerometer data are slower than expected from a normally kinetic person and indicative of bradykinesia based upon a comparative statistical analysis of the determined sets of spectral components of the single movements of peak acceleration that occur over the extended period of time.
-
-
17. A non-transitory computer program product for determining a bradykinetic state of a person, which is a state characterized by slowness of movement relative to a normally kinetic person and in which movements involving peaks of acceleration have strong low frequency components, the computer program product comprising:
-
a processor; a memory containing a computer-executable instruction that configures the processor to; obtain a time series of data in real-time from an accelerometer worn on an extremity of the person, the time series of accelerometer data being representative of movement of the person over an extended period of time, wherein the time series of accelerometer data is obtained while the accelerometer is oriented to be sensitive to at least one of pronation and supination movements of the person and is attached to at least a limb of the person; band pass filter accelerometer data to extract filtered data for a band of interest, wherein the band of interest has a lower end cut off frequency selected to remove DC; extract a plurality of bins of acceleration data from the time series of the filtered accelerometer data; search the data in each of the plurality of bins to identify a time window of peak acceleration, the time window within a given bin comprising at least two samples that capture accelerometer data for a single movement having a peak acceleration relative to other movements observed within the given bin; and process the data in the time windows of peak acceleration to determine; spectral components for each of the single movements having peak acceleration; and an objective measure of the extent to which the single movements having peak acceleration identified within the plurality of bins of acceleration data are slower than expected from a normally kinetic person and indicative of bradykinesia based upon a comparative statistical analysis of the determined sets of spectral components of the single movements of peak acceleration that occur over the extended period of time.
-
Specification