Patient monitoring system
One embodiment provides a sensor module for patient monitoring. The sensor module includes a processor circuitry; a memory circuitry; at least one time of flight (TOF) sensor; a TOF logic; and a monitor logic. The TOF logic is configured to determine a sequence of elevation maps of at least a portion of a patient room. The TOF logic is further configured to at least one of a distance vector and a velocity vector associated with a selected room occupant based, at least in part, on a plurality of elevation maps in the sequence of elevation maps. Each elevation map is determined based, at least in part, on a respective TOF data set captured from the at least one TOF sensor. Each TOF data set is captured, periodically at a time interval. The monitor logic is configured to classify an activity of the selected room occupant as acceptable or unacceptable. The classifying is based, at least in part, on the at least one of the distance vector and the velocity vector.
1 . A sensor module for monitoring of a patient while maintaining patient privacy, the sensor module comprising:
a processor circuitry;
a memory circuitry;
at least one time of flight (TOF) sensor, configured to capture a sequence of data sets of TOF data while positioned in or on a ceiling in the patient room, such that each respective data set is captured periodically at a time interval, the at least one TOF sensor being associated with one or more pixels, wherein the at least one sensor is not configured to identify specific persons;
a TOF logic configured to determine a sequence of elevation maps of at least a portion of a patient room, the TOF logic further configured to determine at least one of a distance vector and a velocity vector associated with a selected room occupant based, at least in part, on a plurality of elevation maps in the sequence of elevation maps, each elevation map determined based, at least in part, on the respective TOF data set of the sequence of data sets; and
a monitor logic configured to classify an activity of the patient as acceptable or unacceptable based, at least in part, on the at least one of the distance vector and the velocity vector of the patient, to determine a number of room occupants based, at least in part, on a selected elevation map, to log the activity of the patient if the activity of the patient is acceptable, to notify a facility management system if the activity of the patient is unacceptable, and to notify the facility management system if the number of room occupants determined is above or below a threshold,
wherein an acceptable activity is any activity which is not an unacceptable activity, and wherein the unacceptable activity is selected from the group consisting of:
(a) the patient shows no movement for a period of time greater than a threshold value,
(b) the distance and/or velocity vectors associated with the patient are greater than a threshold for at least a predefined time period,
(c) the patient spends longer than a threshold period of time in the bathroom,
(d) a second room occupant is on the bed with the patient, and
(e) the patient is resisting a caregiver.
2 . The sensor module according to claim 1 , wherein the monitor logic is trained using a machine learning technique.
3 . The sensor module according to claim 1 , wherein the TOF data set is captured from a plurality of TOF sensors.
4 . The sensor module according to claim 1 , wherein the monitor logic is configured to receive TOF data from a secondary sensor module.
5 . A method of monitoring a patient, comprising:
capturing a sequence of data sets of time of flight (TOF) data, each respective data set captured periodically at a time interval by at least one TOF sensor positioned in or on a ceiling of the patient room, the at least one TOF sensor being associated with one or more pixels, wherein the at least one sensor is not configured to identify specific persons;
determining, by a time of flight (TOF) logic, a sequence of elevation maps of at least a portion of a patient room, each elevation map determined based, at least in part, on the respective TOF data set of the sequence of data sets;
determining, by the TOF logic, at least one of a distance vector and a velocity vector associated with a selected room occupant based, at least in part, on a plurality of elevation maps in the sequence of elevation maps;
classifying, by a monitor logic, an activity of the patient as acceptable or unacceptable based, at least in part, on the at least one of the distance vector and the velocity vector of the patient;
determining, by the monitor logic, a number of room occupants based, at least in part, on a selected elevation map; and
classifying an activity of the patient as acceptable or unacceptable based, at least in part, on the at least one of the distance vector and the velocity vector of the patient, logging the activity of the patient if the activity of the patient is acceptable, notifying, by the monitor logic, a facility management system if the activity of the patient is unacceptable, and notifying the facility management system if the number of room occupants determined is above or below a threshold,
wherein an acceptable activity is any activity which is not an unacceptable activity, and wherein the unacceptable activity is selected from the group consisting of:
(a) the patient and shows no movement for a period of time greater than a threshold value,
(b) the distance and/or velocity vectors associated with the patient are greater than a threshold for at least a predefined time period,
(c) the patient spends longer than a threshold period of time in the bathroom,
(d) a second room occupant is on the bed with the patient, and
(e) the patient is resisting a caregiver.
6 . The method of claim 5 , wherein the monitor logic is trained using a machine learning technique.
7 . The method of claim 5 , wherein the TOF data set is captured from a plurality of TOF sensors.
8 . The method of claim 7 , wherein the plurality of TOF sensors is included in a single sensor module.
9 . A non-transitory computer readable storage device having stored thereon instructions that when executed by one or more processors result in the following operations comprising: the method according to claim 5 .
10 . A system for monitoring of a patient, while maintaining patient privacy, the system comprising:
a primary sensor module; and
at least one secondary sensor module coupled to the primary sensor module,
the primary sensor module comprising:
a processor circuitry,
a memory circuitry,
at least one time of flight (TOF) sensor configured to capture a sequence of data sets of TOF data while positioned in or on a ceiling in the patient room, such that each respective data set is captured periodically at a time interval, the at least one TOF sensor being associated with one or more pixels, wherein the at least one sensor is not configured to identify specific persons,
a TOF logic configured to determine a sequence of elevation maps of at least a portion of a patient room, the TOF logic further configured to determine at least one of a distance vector and a velocity vector associated with a selected room occupant based, at least in part, on a plurality of elevation maps in the sequence of elevation maps, each elevation map determined based, at least in part, on the respective TOF data set, and
a monitor logic configured to classify an activity of the patient as acceptable or unacceptable based, at least in part, on the at least one of the distance vector and the velocity vector of the patient, to determine a number of room occupants based, at least in part, on a selected elevation map, to log the activity of the patient if the activity of the patient is acceptable, and to notify a facility management system if the activity of the patient is unacceptable,
wherein an acceptable activity is any activity which is not an unacceptable activity, and wherein the unacceptable activity is selected from the group consisting of:
(a) the patient and shows no movement for a period of time greater than a threshold value,
(b) the distance and/or velocity vectors associated with the patient are greater than a threshold for at least a predefined time period,
(c) the patient spends longer than a threshold period of time in the bathroom,
(d) a second room occupant is on the bed with the patient, and
(e) the patient is resisting a caregiver.
11 . The system according to claim 10 , wherein the monitor logic is trained using a machine learning technique.
12 . The system according to claim 10 , wherein the TOF data set is captured from a plurality of TOF sensors.
13 . The system according to claim 10 , wherein the monitor logic is configured to receive TOF data from each secondary sensor module.