Method for monitoring a recumbent patient to obtain information on a body position of the patient

1 . A method for monitoring a recumbent patient to obtain information on a body position of the patient, comprising the steps of:

a) emitting electromagnetic waves by a sender arranged in a distance from the patient;

b) receiving electromagnetic waves reflected by the patient by a receiver arranged in a distance from the patient;

c) determining a movement activity level from an amplitude and/or phase of the received electromagnetic waves;

d) classifying the activity level into at least one high activity level class and into at least one low activity level class, wherein gross motor movements of the patient are classified into the at least one high activity level class; and

e) analyzing the amplitude and/or phase of the received electromagnetic waves relating to an activity level classified in the at least one low activity level class to obtain the information on the body position of the patient, wherein the information on the body position of the patient is a signature of a temporal progression of values obtained from the amplitude and/or phase of the received electromagnetic waves relating to the activity level classified in the at least one low activity level class, and the obtained signature is compared to stored signatures obtained for past time intervals, wherein at least some of the stored signatures are assigned to specific body positions;

wherein the electromagnetic waves are radio waves and wherein the sender and receiver are part of a radar system, and

wherein range Doppler maps are generated from the received radio waves and the activity level is obtained from a variance of a velocity obtained from the received electromagnetic waves by calculating an index of dispersion of range-Doppler matrices of consecutive time intervals.

2 . The method as recited in claim 1 , wherein the classified activity levels are assigned to consecutive time intervals.

3 . The method as recited in claim 1 , wherein the assignment of specific body positions to the stored signatures is carried out by using a machine learning method, the machine learning method comprising either of:

i. training data including various captured or simulated signatures and the respective body positions;

ii. a training phase where the body positions are identified by pressure sensors, video data, acoustic information, or data obtained from wearables; and

iii. a training phase where the body positions are manually entered.

4 . The method as recited in claim 1 , wherein the radar system is a wideband radar system.

5 . The method as recited in claim 1 , wherein a two-dimensional position of the received electromagnetic waves is considered when determining the movement activity level and/or when analyzing the received electromagnetic waves to obtain the information on the body position.

6 . The method as recited in claim 4 , wherein an at least two-dimensional range Doppler map is obtained from signals received by at least two of the receivers and in that positions are assigned to detected movements.

7 . The method as recited in claim 5 , wherein information relating to different parts of the patient's body is independently processed.

8 . The method as recited in claim 1 , comprising the further step of analyzing a temporal progression of data obtained from the received electromagnetic waves to identify a signature relating to vital signs of the patient and by obtaining vital sign information from the identified signature.

9 . The method as recited in claim 8 , wherein a position of a chest region of the patient is identified based on a breathing signature in the received electromagnetic waves.

10 . The method as recited in claim 9 , wherein movement points originating from the patient are identified in the received electromagnetic waves and in that all movement points originating from the patient are collected during a first sample period and clustered to obtain one or several clusters in a position space.

11 . The method as recited in claim 10 , wherein each cluster is classified according to its geometrical shape into a torso candidate class or a non-torso candidate class.

12 . The method as recited in claim 10 , wherein the movement points are periodically collected and for each of the clusters a single value is calculated for all the movement points assigned to the respective cluster, the single value representing an activity level in the respective cluster.

13 . The method as recited in claim 12 , wherein the single value is calculated from a number of movement points collected and absolute velocities of each of the movement points.

14 . The method as recited in claim 12 , wherein the single values are collected over a second sample period.

15 . The method as recited in claim 14 , comprising the further step of performing a frequency analysis over the collected single values and classifying each of the clusters into a chest candidate class or a non-chest candidate class.

16 . The method as recited in claim 7 , wherein information relating to the chest region and information relating to other regions of the patient and a positional relationship between the chest region and the other regions are processed to obtain the information on the body position of the patient.

17 . A method for obtaining information for assessing a patient's risk for bedsore, comprising the following steps:

a) obtain information on a body position of the patient using a method as recited in claim 1 at successive points in time;

b) processing the information to obtain at least one of the following:

a frequency of repositioning activities of the patient;

a distribution of body positions assumed by the patient.

18 . The method as recited in claim 1 , wherein a three-dimensional position, of the received electromagnetic waves is considered when determining the movement activity level and/or when analyzing the received electromagnetic waves to obtain the information on the body position.