# US 2021/0264281 A1 · Anatomical position monitoring for bodily pressure ulcers

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0264281 A1
- Document ID: US20210264281A1
- Status: Published application
- Kind: A1
- Application number: 16/800,238
- Filed: 2020-02-25
- Published: 2021-08-26
- Earliest claimed: 2020-02-25
- Assignees: Ventech Solutions, Inc.
- Applicants: Ventech Solutions, Inc.
- Inventors: Ravi Kiran Pasupuleti; Ravi Kunduru
- Web page: https://patentgrep.com/patent/US20210264281A1

## CPC

- G06N 3/084: Backpropagation, e.g. using gradient descent
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- A61B 5/1114: {Tracking parts of the body}
- A61B 5/1126: {using a particular sensing technique}
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/09: Supervised learning
- G16H 20/30: relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
- G16H 40/67: for remote operation
- A61B 5/0507: using microwaves or terahertz waves
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)

## Abstract

A method and system of training a machine learning neural network (MLNN) in monitoring anatomical positioning causing bodily pressure ulcers (BPUs). The method comprises receiving, in a first input layer of the MLNN, from a millimeter wave (mmWave) radar sensing device, mmWave radar point cloud data representing anatomical positions of the medical patient in association with corresponding durations; receiving, in at least a second layer of the MLNN, attendant attribute data for the durations, the first and the at least a second input layers being interconnected with an output layer of the MLNN via at least one intermediate layer; training a MLNN classifier in accordance with a supervised classification that establishes a correlation between a likelihood of formation of BPUs with the mmWave point cloud data and attendant attribute data; and adjusting the initial matrix of weights by backpropagation to increase correlation with the likelihood of formation of BPUs as generated at the output layer.

## Claims

### Claim 1

A method of training a machine learning neural network (MLNN) in monitoring anatomical positioning causing bodily pressure ulcers (BPUs) of a medical patient, the method performed in one or more processors and comprising:
receiving, in a first input layer of the MLNN, from a millimeter wave (mmWave) radar sensing device, mmWave radar point cloud data representing a set of anatomical positions of the medical patient in association with a corresponding set of durations;
receiving, in at least a second layer of the MLNN, attendant attribute data for the corresponding set of durations, the first and the at least a second input layers being interconnected with an output layer of the MLNN via at least one intermediate layer, the at least one intermediate layer configured in accordance with an initial matrix of weights, the first, at least a second, intermediate and output layers of the MLNN being implemented, using the one or more processors, in a memory of the computing device;
training a MLNN classifier in accordance with a supervised classification that establishes a correlation between a likelihood of formation of BPUs as generated at the output layer with the mmWave point cloud data and the attendant attribute data; and
adjusting the initial matrix of weights by backpropagation thereby to increase the correlation with the likelihood of formation of BPUs as generated at the output layer.

### Claim 2 (depends on 1)

The method of claim 1 wherein the backpropogation comprises recursively adjusting the initial matrix of weights in diminishment of an error matrix calculated at the output layer thereby to increase the correlation with the likelihood of BPUs.

### Claim 3 (depends on 1)

The method of claim 1 wherein the mmWave radar point cloud data includes bodily pressure points data relative to a patient support platform between changes in successive ones of the set of anatomical positions in accordance with the corresponding set of durations.

### Claim 4 (depends on 3)

The method of claim 3 wherein the attendant attribute data comprises at least one of patient specific data and environmental attribute data pertaining to the corresponding set of durations.

### Claim 5 (depends on 4)

The method of claim 4 wherein the patient specific data includes at least one of a patient age, weight, blood pressure, height, sex, and pulse rate.

### Claim 6 (depends on 4)

The method of claim 4 wherein the environmental attribute data comprises at least one of a temperature measurement, a humidity measurement, and a category of physical infrastructure in which the medical patient is located.

### Claim 7 (depends on 1)

The method of claim 1 wherein the mmWave radar sensing device comprises a 60 GHz mmWave radar sensing device.

### Claim 8 (depends on 1)

The method of claim 1 further comprising deploying the trained MLNN classifier upon establishing the correlation with the likelihood of formation of BPUs exceeds a 90 percent threshold value.

### Claim 9 (depends on 1)

The method of claim 1 further comprising deploying the trained MLNN classifier upon receiving, in real time from an edge computing device, subsequent mmWave point cloud data at the first input layer and attendant attribute data received in the at least a second input layer, the edge computing device providing a privacy-centric deployment system associated with the subsequent medical patient.

### Claim 10 (depends on 9)

The method of claim 9 further comprising generating, at the output layer, a likelihood of formation of pressure ulcers for the subsequent medical patient.

### Claim 11

A computing system comprising:
one or more processors;
a memory storing a set of instructions, the instructions when executed in the one or more processors causing operations comprising:
receiving, in a first input layer of the MLNN, from a millimeter wave (mmWave) radar sensing device, mmWave radar point cloud data representing a set of anatomical positions of the medical patient in association with a corresponding set of durations;
receiving, in at least a second layer of the MLNN, attendant attribute data for the corresponding set of durations, the first and the at least a second input layers being interconnected with an output layer of the MLNN via at least one intermediate layer, the at least one intermediate layer configured in accordance with an initial matrix of weights, the first, at least a second, intermediate and output layers of the MLNN being implemented, using the one or more processors, in a memory of the computing device;
training a MLNN classifier in accordance with a supervised classification that establishes a correlation between a likelihood of formation of BPUs as generated at the output layer with the mmWave point cloud data and the attendant attribute data; and
adjusting the initial matrix of weights by backpropagation thereby to increase the correlation with the likelihood of formation of BPUs as generated at the output layer.

### Claim 12 (depends on 11)

The computing system of claim 11 wherein the backpropogation comprises recursively adjusting the initial matrix of weights in diminishment of an error matrix calculated at the output layer thereby to increase the correlation with the likelihood of BPUs.

### Claim 13 (depends on 11)

The computing system of claim 11 wherein the mmWave radar point cloud data includes bodily pressure points data relative to a patient support platform between changes in successive ones of the set of anatomical positions in accordance with the corresponding set of durations.

### Claim 14 (depends on 13)

The computing system of claim 13 wherein the attendant attribute data comprises at least one of patient specific data and environmental attribute data pertaining to the corresponding set of durations.

### Claim 15 (depends on 14)

The computing system of claim 14 wherein the patient specific data includes at least one of a patient age, weight, blood pressure, height, sex, and pulse rate.

### Claim 16 (depends on 14)

The computing system of claim 14 wherein the environmental attribute data comprises at least one of a temperature measurement, a humidity measurement, and a category of physical infrastructure in which the medical patient is located.

### Claim 17 (depends on 11)

The computing system of claim 11 wherein the mmWave radar sensing device comprises a 60 GHz mmWave radar sensing device.

### Claim 18 (depends on 11)

The computing system of claim 11 further comprising deploying the trained MLNN classifier upon establishing the correlation with the likelihood of formation of BPUs exceeds a 90 percent threshold value.

### Claim 19 (depends on 18)

The computing system of claim 18 further comprising deploying the trained MLNN classifier upon receiving, in real time from an edge computing device, subsequent mmWave point cloud data at the first input layer and attendant attribute data received in the at least a second input layer, the edge computing device providing a privacy-centric deployment system associated with the subsequent medical patient.

### Claim 20 (depends on 19)

The computing system of claim 19 further comprising generating, at the output layer, a likelihood of formation of pressure ulcers for the subsequent medical patient.
