Quantitative falls risk assessment through inertial sensors and pressure sensitive platform
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-10258257-B2
Publication Date
2019-04-16
Expiration Date
Abstract
A system, method, and apparatus is provided for estimating a risk of falls from pressure sensor data and inertial sensor data. A classifier function may be generated to relate the inertial sensor and the pressure sensor data with the falls risk (e.g., falls risk or prospective falls data) of the person who generated the pressure sensor data and inertial sensor data. The classifier function may be used to predict a person's risk of falls based on inputs of pressure sensor data and inertial sensor data. Separate classifier functions may be generated for men and women, or separate classifier functions may be generated for patients who closed their eyes while data was being collected and for patients who opened their eyes while data was being collected.
Core Innovation
The invention provides a portable falls-risk assessment method and system in which inertial sensor data and pressure sensor data are used together during a standing balance assessment. An inertial sensor is configured to be attached to a lumbar region of a person, and a pressure sensor matrix comprising a plurality of pressure sensors detects pressure exerted by the person on the pressure sensor matrix.
During the assessment, the processor generates center of pressure data based on the received pressure sensor data and then derives a feature value indicative of when the person has fallen during a predetermined interval. The derived feature value is used for training a classifier model so that the classifier model recognizes the feature value based on the received inertial sensor data and the center of pressure data.
The classifier model is configured to generate a prediction risk of future falls based on the center of pressure data and based on the inertial sensor data input into the classifier model, and the processor uses the prediction risk to generate a falls risk estimate. In this way, the system and method use both center of pressure data and inertial sensor data to estimate falls risk and to support reducing the risk of future falls.
The disclosure additionally specifies separate models for eyes open and/or eyes closed conditions and training constraints based on only men or only women, and derives feature values from inertial sensor data and pressure sensor data, including feature values derived from statistical properties and frequency-domain features such as spectral edge frequency and median frequency.
Claims Coverage
The independent claim set covers three core inventions: a method performing a standing balance assessment using lumbar inertial data and pressure-matrix center-of-pressure data, then deriving a fall-indicative feature, training a classifier, and generating a future-falls prediction risk and falls risk estimate; a corresponding system with a lumbar inertial sensor, a pressure sensor matrix, and a processor configured to perform the same assessment/training/prediction workflow; and a method variant that emphasizes recognizing a feature value indicative of when the person has fallen during a predetermined interval and generating prediction risk and a falls risk estimate.
Lumbar inertial sensor and pressure sensor matrix for standing balance assessment
Receiving inertial sensor data from the inertial sensor configured to be attached to a lumbar region of the person, and receiving pressure sensor data from the pressure sensor matrix that comprises a plurality of pressure sensors detecting pressure exerted by the person on the pressure sensor matrix.
Center of pressure data generation from pressure sensor matrix data
Generating center of pressure data for the person based on at least the received pressure sensor data from the standing balance assessment.
Fall-indicative feature value during a predetermined interval
Deriving a feature value indicative of when the person has fallen during a predetermined interval using the processor, based on the received inertial sensor data and the center of pressure data.
Classifier training to predict future falls risk using inertial and center-of-pressure inputs
Training a classifier model to recognize the feature value, wherein the classifier model is configured to generate a prediction risk of future falls based on the center of pressure data and based on the inertial sensor data input into the classifier model.
Falls risk estimate derived from prediction risk to reduce future falls risk
Using the processor to generate a falls risk estimate based on the prediction risk of future falls to reduce the risk of future falls of the person.
System architecture with lumbar inertial sensor, pressure sensor matrix, and processor
An inertial sensor configured to be attached to a lumbar region of a person; a pressure sensor matrix comprising a plurality of pressure sensors configured to sense pressure exerted by the person thereon; and at least one processor configured to perform a standing balance assessment, generate center of pressure data, derive a feature value, train a classifier model, generate a prediction risk of future falls, and use the prediction risk to generate a falls risk estimate.
Across the independent claims, the main inventive elements are the combination of lumbar-mounted inertial sensing with a pressure sensor matrix to generate center of pressure data, deriving a feature value indicative of when the person has fallen during a predetermined interval, training a classifier using inertial and center-of-pressure inputs to generate a prediction risk of future falls, and generating a falls risk estimate intended to reduce the risk of future falls.
Stated Advantages
Reduce the risk of future falls of the person.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?