# US 9,433,389 B2 · Method for monitoring the accuracy of tissue motion prediction from surrogates

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

- Number: US 9,433,389 B2
- Document ID: US9433389B2
- Status: Granted
- Kind: B2
- Application number: 13/531,044
- Filed: 2012-06-22
- Published: 2013-01-17
- Granted: 2016-09-06
- Earliest claimed: 2011-07-12
- Assignees: University of Maryland, Baltimore; University of Maryland, College Park
- Applicants: Warren D'Souza; Kathleen Malinowski; Thomas McAvoy
- Inventors: Warren D'Souza; Kathleen Malinowski; Thomas McAvoy
- Web page: https://patentgrep.com/patent/US9433389B2

## CPC

- A61B 6/12: Arrangements for detecting or locating foreign bodies
- A61B 6/5217: {extracting a diagnostic or physiological parameter from medical diagnostic data}
- A61N 5/1049: {for verifying the position of the patient with respect to the radiation beam}
- G16H 50/50: for simulation or modelling of medical disorders
- G16Z 99/00: Subject matter not provided for in other main groups of this subclass
- A61B 5/113: occurring during breathing
- A61N 5/1037: {taking into account the movement of the target, e.g. 4D-image based planning}
- A61N 5/1067: {in real time, i.e. during treatment}
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- A61N: ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY (measurement of bioelectric currents A61B; surgical instruments, devices or methods for transferring non-mechanical forms of energy to or from the body A61B18/00; anaesthetic apparatus in general A61M; incandescent lamps H01K; infrared radiators for heating H05B)
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16Z: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR

## Abstract

A system and method for indirectly monitoring the position of a target inside a body is disclosed. The method includes generating position data associated with one or more surrogate devices and predicting a location of the target from the position data based on a target position model that establishes a relationship between an actual location of the target and the position data of the one or more surrogate devices. The method also includes determining that the predicted location of the target deviates from the actual location of the target when an analysis of an error prediction model results in a confidence threshold being exceeded.

## Claims

### Claim 1

A method for indirectly monitoring the position of a target inside a body, the method comprising:
generating position data associated with one or more surrogate devices;
predicting a location of the target from the position data based on a target position model that establishes a relationship between an actual location of the target and the position data;
determining, using an analysis of an error prediction model, accuracy of the relationship between an actual location of the target and the position data; and
determining that the predicted location of the target deviates from the actual location of the target when the analysis of the error prediction model results in a confidence threshold being exceeded.

### Claim 2 (depends on 1)

The method recited in claim 1, further comprising delivering a medical treatment to the predicted location of the target.

### Claim 3 (depends on 2)

The method recited in claim 2, wherein said generating position data comprises receiving a signal from the one or more surrogate devices attached to the body and generating the position data from the received signal.

### Claim 4 (depends on 2)

The method recited in claim 2, further comprising
receiving calibration position data and images of the target, wherein the calibration position data and the images are synchronized over a period of time,
determining target site positions of the target from the received images, and
generating the target position model by determining a correlation between the calibration position data and the determined target site positions.

### Claim 5 (depends on 2)

The method recited in claim 2, further comprising, if the predicted location of the target deviates from the actual location of the target, pausing the delivering of the medical treatment, updating the target position model, and further delivering the medical treatment.

### Claim 6 (depends on 2)

The method recited in claim 2, wherein the analysis of the error prediction model includes a statistical analysis that determines whether the relationship between the actual location of the target and the position data has changed.

### Claim 7 (depends on 6)

The method recited in claim 6, wherein the statistical analysis includes a Hotelling and Q-statistic.

### Claim 8 (depends on 2)

The method recited in claim 2, wherein the analysis of the error prediction model includes a statistical analysis that determines an amount of variance in the position data not captured by the target position model.

### Claim 9 (depends on 8)

The method recited in claim 8, wherein the statistical analysis includes an input variable squared prediction error.

### Claim 10 (depends on 2)

The method recited in claim 2, wherein the error prediction model determines whether the predicted location of the target deviates occurs during a treatment period using only the position data.

### Claim 11

A system for indirectly monitoring the position of a target inside a body, the system comprising:
a sensor device configured to provide position data; and
a processor for (i) predicting a location of the target from the position data using a target position model that establishes a relationship between the target and the position data, (ii) determining, using an analysis of an error prediction model, accuracy of the relationship between an actual location of the target and the position data, and (iii) determining that the predicted location of the target has deviated from the actual location of the target when the analysis of the error prediction model results in a confidence threshold being exceeded.

### Claim 12 (depends on 11)

The system recited in claim 11, further comprising a treatment device configured to deliver a medical treatment to the predicted location of the target of the body for a treatment period.

### Claim 13 (depends on 12)

The system recited in claim 12, wherein the sensor device receives a signal from the one or more surrogate devices attached to the patient and generates the position data based on the received signal.

### Claim 14 (depends on 12)

The system as recited in claim 12, wherein, if the predicted location of the target has deviated from the actual location of the target, the processor controls the treatment device to stop delivering the medical treatment, updates the target position model, and controls the treatment device to start delivering the medical treatment again.

### Claim 15 (depends on 12)

The system recited in claim 12, wherein the analysis includes a statistical analysis that determines if the relationship between the actual location of the target and the position data has changed.

### Claim 16 (depends on 15)

The system recited in claim 15, wherein the statistical analysis includes a Hotelling and Q-statistic.

### Claim 17 (depends on 12)

The system recited in claim 12, wherein the analysis includes a statistical analysis that determines an amount of variance in the position data not captured by the target position model.

### Claim 18 (depends on 17)

The system recited in claim 17, wherein the statistical analysis includes an input variable squared prediction error.

### Claim 19 (depends on 12)

The system recited in claim 12, wherein the processor uses the error prediction model to determine whether the predicted location of the target has deviated during the treatment period using only the position data.

### Claim 20

A method of indirectly monitoring the position of a target inside a body, the method comprising:
generating a target position model that predicts a location of a target of a patient and generating an error prediction model used to determine accuracy of the predicted location of the target;
generating position data based on the respective positions of one or more surrogate devices coupled to the patient;
predicting a location of the target using the target position model and the position data;
delivering a medical treatment to the patient based on the target position model during a treatment period;
determining whether the predicted location of the target has deviated from the actual location of the target using only the position data and the error prediction model during the treatment period; and
when the predicted location of the target has deviated from the actual location of the target, pausing delivering of the medical treatment, updating the target prediction model, and resuming the delivering of the medical treatment.