Systems and methods for fall detection
Inventors
Dohrmann, Anthony • Keeley, David W. • Salcido, Roberto Abel • Mitchell, James
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12011259-B2
Publication Date
2024-06-18
Expiration Date
Abstract
A system for monitoring and detecting the gait and other health related parameters of a user. One such parameter is monitoring of medication compliance and treatment session attendance done by a medication and liquid dispensing apparatus, which combines mechanical dispensing of medication. These parameters are provided in standard of care summaries to care providers, and are continually reported by the Optimum Recognition Blueprint as Standard of Care Summaries to care providers, as well as communicated to the end-user by the Virtual Caregiver Interface.
Core Innovation
The invention provides an electronic caregiver system and virtual system for automated fall detection and reporting using a depth camera configured to be pointed at a human. The virtual system receives a first data set of virtual movement markers comprising a cluster of data points in a shape of a body of the human, including a first movement marker estimated to a head location, a second movement marker estimated to a spine location, and a third movement marker estimated to a joint location.
The virtual system also receives a second data set indicating volume in an area of a location of the human and detects a first irregular pattern of activity based on the volume in the area over a prescribed period of time. An alert is sent to a reporting device indicating that the human has fallen based on the detection of the first irregular pattern of activity, and a second irregular pattern of activity is detected from changes in the first data set of virtual movement markers, with a second alert sent based on the second detected irregular pattern.
Fall detection further includes detecting a rapid acceleration of at least two movement markers comprising the first data set, followed by a rapid deceleration, where at least one of the rapid acceleration and the rapid deceleration includes a rapid change in direction of a location of at least one movement marker. The invention can process the received depth-camera data as part of an Optimum Recognition Blueprint and communicate alerts toward a reporting device and optionally to a monitoring central station for dispatch of emergency services, using a standardized-care summary and a Virtual Caregiver Interface to communicate compliance and treatment status and other health indicators.
Claims Coverage
The independent claims cover a virtual method, a corresponding system, and a non-transitory processor-readable medium implementing automated fall detection and reporting. Across the independent claims, the core coverage includes detecting irregular patterns of activity from depth-camera-derived virtual movement markers and volume in an area, and confirming fall indication using accelerometer-detected rapid acceleration and rapid deceleration with rapid change in direction. In total, three independent claim formulations are explicitly identified: clm-00001, clm-00010, and clm-00018.
Depth-camera virtual movement markers and volume-based irregular activity detection
Receiving from a depth camera a first data set of virtual movement markers comprising a cluster of data points in a shape of a body of the human (estimated head, spine, and joint movement markers) and receiving from the depth camera a second data set indicating volume in an area of a location of the human; detecting a first irregular pattern of activity for the human based on the volume in the area over a prescribed period of time and sending a first alert to a reporting device indicating that the human has fallen based on the detection of the first irregular pattern of activity; detecting a second irregular pattern of activity for the human from changes in the first data set of virtual movement markers and sending a second alert indicating that the human has fallen based on the second detected irregular pattern of activity.
Accelerometer-confirmed rapid acceleration, rapid deceleration, and rapid change in direction
Detecting, from an accelerometer, a rapid acceleration of at least two of the movement markers comprising the first data set of virtual movement markers, followed by a rapid deceleration of the at least two of the movement markers, wherein at least one of the rapid acceleration and the rapid deceleration includes a rapid change in direction of a location of at least one of the movement markers comprising the first data set of virtual movement markers.
Optimum Recognition Blueprint processing for depth-camera data
Receiving and processing the first data set and the second data set from the camera against an optimum recognition blueprint.
Alert communications to a monitoring central station for emergency dispatch
Sending the first and second alerts to a monitoring central station containing information needed to dispatch local emergency services to the location of the human.
Head location represented by initial depth, vertical, and horizontal positions
Determining that a human head location includes initial depth position, initial vertical position, and initial horizontal position.
Rapid acceleration detection via head position component changes
Detecting rapid acceleration of at least two movement markers by detecting changes in the head’s initial depth position, initial vertical position, and initial horizontal position.
Predefined threshold-based alert transmission using time-measured head position changes
Transmitting an alert to a reporting device when a predefined threshold is exceeded based on time-measured changes in an initial depth position of a head, an initial vertical position of the head, and/or an initial horizontal position of the head.
Across the independent claim set, the invention covers automated fall detection and reporting by combining depth-camera-derived virtual movement markers and volume-based irregular activity patterns with accelerometer-detected rapid acceleration and rapid deceleration including rapid change in direction, producing multiple alerts based on detected irregular patterns.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?