Systems and methods for dynamically identifying a patient support surface and patient monitoring

Inventors

Clark, Mattdel Carpio, DerekChapman, Kenneth

Assignees

CareView Communications Inc

Interested in licensing this patent?

MTEC can help explore whether this patent might be available for licensing for your application.

Publication Number

US-10390738-B2

Patent

Publication Date

2019-08-27

Expiration Date


Abstract

Various patient monitoring systems can include a sensor configured to collect three dimensional information. The systems can identify a location of a patient support surface based on the three dimensional information. The systems can set a two dimensional planar threshold based on the patient support surface. The systems can identify a patient location above the patient support surface based on the three dimensional information and compare the patient location to the two dimensional planar threshold. Exceeding the threshold can be indicative of a high risk of a patient fall. An alert can be generated based on the threshold being exceeded. The systems can repeat the identification of the patient support surface location and the setting of the threshold to account for changes in the patient area.

Core Innovation

A patient monitoring system monitors a patient area using sensor output that provides a plurality of frames containing three dimensional information. The three dimensional information includes grid positions of pixels represented by coordinates along X- and Y-axes and depth of each pixel relative to a floor along a Z-axis, and the control circuitry receives the three dimensional information and processes the frames as a chronological series.

The control circuitry identifies a location of a patient support surface by segmenting the three dimensional information into candidate areas and grouping one or more planar surfaces in the candidate areas based on pixel depth. The system sets at least one threshold above the location of the patient support surface on the Z-axis, identifies one or more patient locations based on pixels that are above a depth of the patient support surface and within the location of the patient support surface, and compares the one or more patient locations to the at least one threshold.

The control circuitry generates an output for display on a user interface based on the one or more patient locations traversing the at least one threshold. The system repeats identifying the location of the patient support surface and re-setting the at least one threshold to dynamically adjust the threshold in response to changes in the location of the patient support surface, and uses frame chronology and buffering to compute nominal locations and to reduce errors caused by scene changes, temporary occlusions, and noise.

Claims Coverage

The independent claims (system, method, and computer-readable media) each include a core set of inventive features: receiving 3D frame data, segmenting and grouping planar surfaces to identify a patient support surface, setting a Z-axis threshold above that surface, identifying patient locations from pixels above the support-surface depth within the support-surface location, comparing those locations to the threshold, and generating a user interface output when the locations traverse the threshold. Three additional inventive features recur across the claims: boundary definition, dynamic threshold adjustment, and spatial or pixel-based constraints on patient-location identification.

Receiving three dimensional information of a plurality of frames

The system receives the three dimensional information of the plurality of frames, where the three dimensional information comprises grid positions of a plurality of pixels represented by coordinates along X- and Y-axes and depth of each pixel relative to a floor along a Z-axis.

Identifying a location of a patient support surface by segmenting and grouping planar surfaces

The system identifies a location of a patient support surface by segmenting the three dimensional information into candidate areas and grouping one or more planar surfaces in the candidate areas based on pixel depth.

Setting a Z-axis threshold above the patient support surface

The system sets at least one threshold above the location of the patient support surface on the Z-axis.

Identifying patient locations based on pixels above the support-surface depth within the support-surface location

The system identifies one or more patient locations based on pixels that are above a depth of the patient support surface and within the location of the patient support surface.

Comparing patient locations to the Z-axis threshold and generating a user-interface output for threshold traversal

The system compares the one or more patient locations to the at least one threshold and generates an output for display on a user interface based on the one or more patient locations traversing the at least one threshold.

Defining support-surface boundaries based on height contrast to the floor

The system determines boundaries corresponding to the location of the patient support surface based on a contrast in height from a given surface to the floor using the 3D information.

Grouping non-coplanar planar surfaces

The system groups non-coplanar ones of the one or more planar surfaces.

Dynamically re-setting the Z-axis threshold upon support-surface changes

The system repeats identifying the location of the patient support surface and re-setting the at least one threshold to dynamically adjust the threshold in response to changes in the location of the patient support surface.

Extending the at least one threshold laterally along the X- and Y-axes within the support-surface location

The system sets the at least one threshold to extend laterally along the X- and Y-axes within the location of the patient support surface.

Selecting patient locations using pixels associated with bodily features

The system identifies one or more patient locations by analyzing pixels associated with the bodily features of the patient.

Threshold-based pixel quantity criterion for identifying patient locations

The method identifies one or more patient locations by grouping image pixels that exceed a threshold amount and that are above the depth of the patient support surface within the patient support surface area.

Across the independent claims, the inventive coverage centers on identifying the patient support surface through segmentation into candidate areas and grouping planar surfaces by pixel depth, then setting a Z-axis threshold above that surface to select patient locations from pixels whose depth rises above the support-surface depth within the support-surface location, and generating a user-interface output based on threshold traversal. Dependent claims further narrow planar-surface grouping, boundary definition via height contrast to the floor, dynamic threshold adjustment with repeated support-surface identification, and additional spatial or selection constraints for patient-location identification.

Stated Advantages

Generates an output for display on a user interface based on patient locations traversing the at least one threshold.

Dynamically adjusts the threshold in response to changes in the location of the patient support surface.

Reduces errors from scene changes, temporary occlusions, and noise by using frame chronology and buffering to compute nominal locations.

Documented Applications

Monitoring a patient area by identifying patient locations relative to a patient support surface using three dimensional information and generating an output for display on a user interface based on threshold traversal.

Using the system’s threshold traversal output to indicate increased fall risk.

JOIN OUR MAILING LIST

Stay Connected with MTEC

Keep up with active and upcoming solicitations, MTEC news and other valuable information.