Radiation treatment based upon user specification of at least one custom DVH estimation model training feature

Patent metadata

Field	Value
Publication number	US-10653893-B2
Application number	US-201715690525-A
Country	US
Kind code	B2
Filing date	Aug 30, 2017
Priority date	Aug 30, 2017
Publication date	May 19, 2020
Grant date	May 19, 2020

Abstract

A control circuit provides an opportunity via a user interface for a user to specify at least one custom DVH estimation model training feature. The control circuit then combines a predetermined set of DVH estimation model training features with a user-specified customer DVH estimation model training feature to provide a combined feature set. The control circuit uses the combined feature set to train a knowledge-based DVH estimation model which is then used to provide a DVH estimation for use when developing/optimizing a radiation treatment plan. That resultant radiation treatment plan then controls a radiation-administration platform to provide a therapeutic radiation dose to a patient.

First claim

What is claimed is:

1. An apparatus comprising:
a memory having a radiation treatment plan stored therein and a predetermined set of dose volume histogram (DVH) estimation model training features;
a user interface;
a control circuit operably coupled to the memory and the user interface and configured to:
provide an opportunity via the user interface for a user to specify at least one custom DVH estimation model training feature;
combine the predetermined set of DVH estimation model training features with a user-specified custom DVH estimation model training feature to provide a combined feature set;
use the combined feature set to train a knowledge-based DVH estimation model;
use the trained knowledge-based DVH estimation model to provide a DVH estimation for the radiation treatment plan;
use the radiation treatment plan to control a radiation-administration platform to provide a therapeutic radiation dose to a patient.

2. The apparatus of claim 1 wherein the control circuit is configured to provide the opportunity to specify the at least one custom DVH estimation model training feature via a scripting-based opportunity.

3. The apparatus of claim 1 wherein the control circuit is configured to provide the opportunity to specify the at least one custom DVH estimation model training feature by providing an opportunity to specify a new model statistic.

4. The apparatus of claim 3 wherein the new model statistic comprises a statistic regarding a distance.

5. The apparatus of claim 1 wherein the control circuit is configured to provide the opportunity to specify the at least one custom DVH estimation model training feature by providing an opportunity to specify a radiation beam metric.

6. The apparatus of claim 5 wherein the radiation beam metric comprises a weighted geometric metric for a corresponding radiation beam direction.

7. The apparatus of claim 1 wherein the control circuit is configured to provide the opportunity to specify the at least one custom DVH estimation model training feature by providing an opportunity to specify at least one physical difference between an application setting to be modeled and a given reference application setting.

8. The apparatus of claim 1 wherein the control circuit is further configured to:
use substitute tissue information as a basis for training the knowledge-based DVH estimation model instead of organ at risk (OAR) information.

9. The apparatus of claim 8 wherein the control circuit is configured to use substitute tissue information by selectively avoiding tissue that includes bone.

10. The apparatus of claim 8 wherein the control circuit is configured to use substitute tissue information by selectively including tissue that includes bone.

11. The apparatus of claim 8 wherein the substitute tissue information corresponds to substitute tissue that is within a predetermined degree of similarity to density of the OAR.

12. A method comprising:
providing a memory having a radiation treatment plan stored therein and a predetermined set of dose volume histogram (DVH) estimation model training features;
providing a user interface;

by a control circuit operably coupled to the memory and the user interface:
providing an opportunity via the user interface for a user to specify at least one custom DVH estimation model training feature;
combining the predetermined set of DVH estimation model training features with a user-specified custom DVH estimation model training feature to provide a combined feature set;
using the combined feature set to train a knowledge-based DVH estimation model; using the trained knowledge-based DVH estimation model to provide a DVH estimation for the radiation treatment plan;
using the radiation treatment plan to control a radiation-administration platform to provide a therapeutic radiation dose to a patient.

13. The method of claim 12 wherein providing the opportunity to specify the at least one custom DVH estimation model training feature comprises using a scripting-based opportunity.

14. The method of claim 12 wherein providing the opportunity to specify the at least one custom DVH estimation model training feature comprises providing an opportunity to specify a new model statistic.

15. The method of claim 14 wherein the new model statistic comprises a statistic regarding a distance.

16. The method of claim 12 wherein providing the opportunity to specify the at least one custom DVH estimation model training feature comprises providing an opportunity to specify a radiation beam metric.

17. The method of claim 16 wherein the radiation beam metric comprises a weighted geometric metric for a corresponding radiation beam direction.

18. The method of claim 1 further comprising:
using substitute tissue information as a basis for training the knowledge-based DVH estimation model instead of organ at risk (OAR) information.

19. The method of claim 18 wherein the substitute tissue information corresponds to substitute tissue that is within a predetermined degree of similarity to density of the OAR.

Assignees

• Varian Medical Systems Int Ag

Inventors

• Kuusela Esa
• Cordero Marcos Maria
• Laaksonen Hannu

Classifications

• A61N5/1048 Primary

  Monitoring, verifying, controlling systems and methods · CPC title

• A61N5/1031 Primary

  using a specific method of dose optimization · CPC title

• A61N2005/1074

  Details of the control system, e.g. user interfaces · CPC title