# US 11,908,582 B2 · Assessing the functional performance of an individual

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,908,582 B2
- Document ID: US11908582B2
- Status: Granted
- Kind: B2
- Application number: 17/416,400
- Filed: 2019-12-19
- Published: 2022-02-24
- Granted: 2024-02-20
- Earliest claimed: 2018-12-21
- Assignees: Administracion General De La Comunidad Autonoma De Euskadi
- Applicants: AdministracióN General De La Comunidad AutóNoma De Euskadi
- Inventors: Jaime Herran Planchuelo; Larraitz Añorga Gomez; Itziar Vergara Micheltorena; Kalliopi Vrotsou; Marcos Jesus Arauzo Bravo; Ander Matheu Fernández
- Web page: https://patentgrep.com/patent/US11908582B2

## CPC

- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G16H 50/30: for calculating health indices; for individual health risk assessment
- A61B 5/1114: {Tracking parts of the body}
- A61B 5/1118: {Determining activity level}
- A61B 5/7257: {using Fourier transforms}
- A61B 2562/0219: Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
- A61B 2562/046: in a matrix array
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)

## Abstract

The description relates to a system for assessing the functional performance of an individual, the system comprising a sensor device; a monitoring system, connectable to the sensor device, configured to receive signals coming from the sensor device, pre-process the signals received from the sensor device, extract the main features of the pre-processed signals to be used as predictor variables, assess the functional performance of an individual based on the predictor variables.

## Claims

### Claim 1

A system for assessing functional performance of an individual based on an analysis of a process of sitting down and standing up performed by the individual, the system comprising:
a sensor device, wherein the sensor device is a distributed pressure sensor, wherein the sensor device comprises an array of sensors of Nr×Nc that produces a Nr×Nc measurement matrix for each sampling time;
a monitoring system, connectable to the sensor device, configured to:
receive signals coming from the sensor device, wherein received signals are in the form of a tri-dimensional tensor TS of order Nr×Nc×NS, where Nr is a number of rows of array of sensors, Nc is the number of columns of array of sensors, and NS is a number of sampling times;
pre-process the signals received from the sensor device, wherein pre-processing, by the monitoring system, the signals received from the sensor device comprises:
producing, by the monitoring system, an equi-order tensor TMAX of dimensions Nr×Nc×NMAX;
vectoring the produced equi-order tensor TMAX into a vector V of length NMAX, in which to each position i in is assigned a sum of the elements of the Nr×Nc measurement matrix in a position i of the tensor TMAX;
extract main features of the pre-processed signals as extracted predictor variables; and
assess the functional performance of an individual based on the extracted predictor variables.

### Claim 2 (depends on 1)

The system according to claim 1, wherein producing, by the monitoring system, the equi-order tensor TMAX of dimensions Nr×Nc×NMAX comprises:
reformatting the tri-dimensional tensor TS of order Nr×Nc×NS according to:
if NS≥NMAX, truncating a third dimension of the tensor TS to the length NMAX, transforming the tri-dimensional tensor TS into the equi-order tensor TMAX of dimensions Nr×Nc×NMAX;
if NS≥NMAX, padding, by the monitoring system, the third dimension of the tri-dimensional tensor TS to the length NMAX, transforming the tri-dimensional tensor TS into the equi-order tensor TMAX of dimensions Nr×Nc×NMAX; and
if NS=NMAX, renaming the tri-dimensional tensor TS as the equi-order tensor TMAX.

### Claim 3 (depends on 2)

The system according to claim 2, wherein padding, by the monitoring system, the third dimension of the tri-dimensional tensor TS to the length NMAX, if NS≥NMAX, comprises:
attaching after the tri-dimensional tensor TS of order Nr×Nc×NS a padding tensor Tp of dimensions Nr×Nc×Np, where Np is a padding length Np=NMAX−NS.

### Claim 4 (depends on 1)

The system according to claim 1, wherein the monitoring system extracts, the main features of the pre-processed signals to be used as the extracted predictor variables, the extracted predictor variables being signal descriptive variables in the time domain by:
obtaining the signal descriptive variables in a time domain vector VDESCRIPTIVE.

### Claim 5 (depends on 4)

The system according to claim 4, wherein the signal descriptive variables in the time domain vector VDESCRIPTIVE are selected from at least one of the following:
a range of signal variation;
an instant of signal maximum peak;
a signal standard deviation;
a signal slope defined as a ratio between the range of signal variation and the instant of the signal maximum peak;
a start time of signal change; and
a signal integral.

### Claim 6 (depends on 1)

The system according to claim 1, wherein pre-processing, by the monitoring system, the signals received from the sensor device comprises:
correcting, by the monitoring system, the background of the vector V to obtain a background corrected vector Vb.

### Claim 7 (depends on 6)

The system according to claim 6, wherein correcting, by the monitoring system, the background of the vector V comprises:
subtracting from each element of the vector V the value of a minimum element of V, VMIN to obtain the background corrected vector Vb.

### Claim 8 (depends on 7)

The system according to claim 7, wherein the monitoring system extracts the main features of the pre-processed signals to be used as the extracted predictor variables, the extracted predictor variables being signal time series variables in the time domain by:
filtering the noise of the background corrected vector with a smoothing moving window to obtain a filtered background corrected vector; and
reducing, by the monitoring system, a sample length of the filtered background corrected vector by a decimation factor of d, to obtain the signal time series variables in a time domain vector VTIME.

### Claim 9 (depends on 8)

The system according to claim 8, wherein reducing, by the monitoring system, the sample length of the filtered background corrected vector by the decimation factor of d comprises the following:
applying a simple resampling of the filtered background corrected vector at each time d;
applying a lowpass Chebyshev Type I Infinite Impulse Response (IIR) filter; and
applying a Finite Impulse Response (FIR) filter with a Hamming window.

### Claim 10 (depends on 8)

The system according to claim 8, wherein the monitoring system extracts the main features of the pre-processed signals to be used as the extracted predictor variables, the extracted predictor variables being signal spectral variables in the frequency domain by:
truncating the background corrected vector Vb length to a floor operator LT of maximum power of 2 of the length of the background corrected vector Vb to obtain a truncated background corrected vector;
applying a Fast Fourier transform to the truncated background corrected vector using a Discrete Fast Fourier method;
obtaining the spectral power P of the background corrected vector Vb based on the result of applying the Fast Fourier transform to the truncated background corrected vector; and
truncating the obtained spectral power P to obtain the extracted predictor variables as signal spectral variables in a frequency domain vector VFREQUENCY is obtained.

### Claim 11 (depends on 10)

The system according to claim 10, wherein the monitoring system extracts the main features of the pre-processed signals to be used as the extracted predictor variables by:
obtaining the respective maximum and minimum values of the extracted predictor variables in one or more of the vectors VDESCRIPTIVE, VTIME and/or VFREQUENCY,
normalizing, based on the obtained maximum and minimum values, between 0 and 1, the extracted predictor variables in one or more of the vectors VDESCRIPTIVE, VTIME and/or VFREQUENCY to obtain normalized extracted predictor variables;
attaching in a vector VP the normalized extracted predictor variables.

### Claim 12 (depends on 1)

The system according to claim 1, wherein assessing, by the monitoring system, the functional performance of an individual based on the extracted predictor variables comprises:
predicting a value of a TUG (Timed Up Go) test of the individual based on the extracted predictor variables; and
assessing the functional performance of an individual based on the predicted value of the TUG test.

### Claim 13 (depends on 1)

The system according to claim 1, wherein the monitoring system assesses the functional performance of an individual based on the extracted predictor variables by:
identifying a state of fragility or robustness of the individual based on the extracted predictor variables.

### Claim 14 (depends on 1)

The system according to claim 1, wherein the monitoring system assesses the functional performance of an individual based on the extracted predictor variables by:
identifying individuals with high risk of fractures and other health related adverse outcomes based on the extracted predictor variables.