Artificial intelligence boosting dose calculations

Patent metadata

Field	Value
Publication number	US-12406756-B2
Application number	US-202217692091-A
Country	US
Kind code	B2
Filing date	Mar 10, 2022
Priority date	Mar 10, 2022
Publication date	Sep 2, 2025
Grant date	Sep 2, 2025

Abstract

Embodiments described herein provide for training an artificial intelligence model to boost dose depositions. The artificial intelligence model receives medical images and a dose deposition determined according to a first dose deposition model. The artificial intelligence model modifies the received dose deposition determined according to the first dose deposition model such that the dose deposition simulates a dose deposition determined by a second dose deposition model.

First claim

What we claim is:

1. A method of improving dose predictions of a first dose deposition model using an artificial intelligence model thereby enhancing radiation treatment for a patient, the method comprising:
receiving, by a processor, a medical image of an anatomical region of a patient from a database or a radiotherapy machine through a network;
determining, by the processor, an initial dose deposition of the anatomical region using the medical image and determined by the first dose deposition model indicating a minimum dose deposition;
determining, by the processor, an adjusted dose deposition for the medical image corresponding to an increased accuracy of the initial dose deposition using the medical image and determined by a second dose deposition model, wherein determining the adjusted dose deposition comprises removing one or more artifacts corresponding to the medical image;
training, by the processor, the artificial intelligence model to determine simulated adjusted dose depositions using the initial dose deposition, the adjusted dose deposition, and a set of training medical images, each training medical image having a corresponding first training dose deposition determined by the first dose deposition model and an adjusted training dose deposition determined by the second dose deposition model;
generating, by the processor, a simulated adjusted dose deposition by applying the trained artificial intelligence model; and
adjusting, by the processor, at least one configuration radiation parameters of the radiotherapy machine based on the simulated adjusted dose deposition for the patient.

2. The method of claim 1 , wherein the first dose deposition model determines a flux distribution, abstracting a dose reaction rate with the anatomical region of the patient.

3. The method of claim 1 , wherein the second dose deposition model employs a nondeterministic particle behavior simulator and/or direct dose measurements.

4. The method of claim 1 , wherein the minimum dose deposition determined using the first dose deposition model is adjusted using the artificial intelligence model.

5. The method of claim 1 , wherein the artificial intelligence model is a neural network.

6. The method of claim 1 , wherein the set of training medical images and corresponding first training dose deposition determined by the first dose deposition model and the adjusted training dose deposition determined by the second dose deposition model corresponds to a particular anatomical region.

7. The method of claim 1 , wherein the set of training medical images and corresponding first training dose deposition determined by the first dose deposition model and the adjusted training dose deposition determined by the second dose deposition model corresponds to a particular clinician.

8. The method of claim 1 , wherein the set of training medical images and corresponding first training dose deposition determined by the first dose deposition model and the adjusted training dose deposition determined by the second dose deposition model corresponds to a particular radiotherapy machine.

9. The method of claim 1 , wherein the adjusted training dose deposition is modified according to a preference of a medical professional, the artificial intelligence model being trained to simulate the adjusted training dose deposition determined by the second dose deposition model and modified according to the preference of the medical professional.

10. A system for improving dose predictions of a first dose deposition model using an artificial intelligence model thereby enhancing radiation treatment for a patient, the system comprising:
a server comprising a processor and a non-transitory computer-readable medium containing instructions that when executed by the processor cause the processor to perform operations comprising:
receiving a medical image of an anatomical region of a patient from a database or a radiotherapy machine through a network;
determining an initial dose deposition of the anatomical region using the medical image and determined by the first dose deposition model indicating a minimum dose deposition;
determining an adjusted dose deposition for the medical image corresponding to an increased accuracy of the initial dose deposition using the medical image and determined by a second dose deposition model, wherein determining the adjusted dose deposition comprises removing one or more artifacts corresponding to the medical image;
training the artificial intelligence model to determine simulated adjusted dose depositions using the initial dose deposition, the adjusted dose deposition, and a set of training medical images, each training medical image having a corresponding first training dose deposition determined by the first dose deposition model and an adjusted training dose deposition determined by the second dose deposition model;
generating a simulated adjusted dose deposition by applying the trained artificial intelligence model; and
adjusting at least one configuration radiation parameters of the radiotherapy machine based on the simulated adjusted dose deposition for the patient.

11. The system according to claim 10 , wherein the first dose deposition model determines a flux distribution abstracting a dose reaction rate with an anatomical region of the patient.

12. The system according to claim 10 , wherein the second dose deposition model employs a nondeterministic particle behavior simulator and/or direct dose measurements.

13. The system according to claim 10 , wherein the minimum dose deposition determined using the first dose deposition model is adjusted using the artificial intelligence model.

14. The system according to claim 10 , wherein the artificial intelligence model is a neural network.

15. The system according to claim 10 , wherein the set of training medical images and corresponding first training dose deposition determined by the first dose deposition model and the adjusted training dose deposition determined by the second dose deposition model corresponds to a particular anatomical region.

16. The system according to claim 10 , wherein the set of training medical images and corresponding first training dose deposition determined by the first dose deposition model and the adjusted training dose deposition determined by the second dose deposition model corresponds to a particular clinician.

17. The system according to claim 10 , wherein the set of training medical images and corresponding first training dose deposition determined by the first dose deposition model and the adjusted training dose deposition determined by the second dose deposition model corresponds to a particular radiotherapy machine.

18. The system according to claim 10 , wherein the adjusted training dose deposition is modified according to a preference of a medical professional, the artificial intelligence model being trained to simulate the adjusted training dose deposition determined by the second dose deposition model and modified according to the preference of the medical professional.

Assignees

• Siemens Healthineers Int Ag

Inventors

• Maslowski Alexander
• Barnett Douglas Allen
• Wareing Todd

Classifications

• G06N3/08

  Learning methods · CPC title

• G16H30/40

  for processing medical images, e.g. editing · CPC title

• G16H50/70

  for mining of medical data, e.g. analysing previous cases of other patients · CPC title

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• G16H50/50

  for simulation or modelling of medical disorders · CPC title