Systems and methods for dose calculation in generating radiation treatment plans

1. A computing system comprising:

a central processing unit (CPU); and

memory coupled to the CPU and having stored therein instructions that, when executed by the computing system, cause the computing system to execute operations associated with generating a radiation treatment plan, the operations comprising:

accessing a first dose calculated using first set of fluence maps and a first dose prediction model;

accessing a second fluence map;

projecting the second fluence map onto the first set of fluence maps to determine a set of scalars and a residual value; and

when the residual value satisfies a criterion, calculating a second dose using the first dose prediction model, the set of scalars, and the second fluence map, and wherein when the residual value does not satisfy the criterion, calculating the second dose using a second dose prediction model and the second fluence map.

2. The computing system of claim 1 , wherein the operations further comprise accessing a set of third doses calculated using the first set of fluence maps and the second dose prediction model;

wherein said calculating a second dose using the first dose prediction model, the set of scalars, and the second fluence map comprises:

calculating a fourth dose using the first dose prediction model and the residual value; and

summing the fourth dose and products of multiplying the set of scalars and the set of third doses.

3. The computing system of claim 1 , wherein the operations further comprise accessing a set of third doses comprising differences between doses calculated using the first set of fluence maps and the first dose prediction model and doses calculated using the first set of fluence maps and the second dose prediction model;

wherein said calculating a second dose using the first dose prediction model, the set of scalars, and the second fluence map comprises:

calculating a fourth dose using the second fluence map and the first dose prediction model; and

summing the fourth dose and products of multiplying the set of scalars and the third set of doses.

4. The computing system of claim 1 , wherein the criterion comprises a threshold value or magnitude of the residual value.

5. The computing system of claim 1 , wherein the criterion comprises a threshold value for a normalized value of the residual value.

6. The computing system of claim 1 , wherein the second dose prediction model is more accurate and utilizes more computational resources than the first dose prediction model.

7. The computing system of claim 1 , wherein the operations further comprise;

selecting a final fluence map for radiation treatment, wherein the final fluence map is selected from a plurality fluence maps comprising at least the first set of fluence maps and the second fluence map; and

converting the final fluence map into leaf sequence settings that control movement of a multileaf collimator during the radiation treatment.

8. A computing system comprising:

a central processing unit (CPU); and

memory coup ed to the CPU and having stored therein instructions that, when executed by the computing system, cause the computing system to execute operations associated with generating a radiation treatment plan, the operations comprising:

calculating first doses using a first set of fluence maps and a first dose prediction model;

accessing a second fluence map;

projecting the second fluence map onto the first set of fluence maps to determine a set of scalars and a residual value; and

selecting, based on the residual value, a dose prediction model to calculate a second dose using the second fluence map, wherein when the residual value satisfies a criterion, then the second dose is calculated using the first dose prediction model and wherein when the residual value does not satisfy the criterion, then the second dose calculated using a second dose prediction model, wherein the second dose prediction model calculates the second doses more accurately than the first dose prediction model.

9. The computing system of claim 8 , wherein the operations further comprise calculating third doses using the first set of fluence maps and the second dose prediction model, wherein further when the residual value satisfies the criterion then the second dose is calculated by:

calculating a fourth dose using the first dose prediction model and the residual value; and

summing the fourth dose and products of multiplying the set of scalars and the third doses.

10. The computing system of claim 8 , wherein the operations further comprise calculating third doses comprising differences between doses calculated using the first set of fluence maps and the first dose prediction model and doses calculating using the first set of fluence maps and the second dose prediction model wherein further when the residual value satisfies the criterion then the second doses are calculated by:

calculating a fourth dose using the second fluence map and the first dose prediction model; and

summing the fourth dose and products of multiplying the set of scalars and the third doses.

11. The computing system of claim 8 , wherein the criterion comprise a threshold value for a magnitude of the residual value.

12. The computing system of claim 8 , wherein the criterion comprises a threshold value for a normalized value of the residual value.

13. The computing system of claim 8 , wherein the operations further comprise:

selecting a final fluence map for radiation treatment, wherein the final fluence map is selected from a plurality of fluence maps comprising at least the first set of fluence maps and the second fluence map; and

converting the final fluence map into leaf sequence settings that control movement of a multileaf collimator during the radiation treatment.

14. A computer-implemented method for radiation treatment planning, the method comprising:

generating a first radiation treatment plan comprising a first set of fluence map, wherein the first radiation treatment plan is generated using a first model that calculates first doses using the first set of fluence maps;

generating a second radiation treatment plan comprising a second fluence map, wherein the second radiation treatment plan is generated by:

projecting the second fluence map onto the first, set of fluence maps to determine a set of scalars d a residual value; and

when the residual value satisfies a criterion, using the first model to calculate a second dose using the set of scalars and the second fluence map, and when the residual value does not satisfy the criterion, using a second model to calculate the second dose using the second fluence map; and

selecting a final radiation treatment plan from a plurality of radiation treatment plans, wherein the plurality of radiation treatment plans comprises at least the first and second radiation treatment plans.

15. The method of claim 14 , further comprising, accessing third doses calculated using the first set of fluence maps and the second dose prediction model;

wherein calculating the second dose using the first dose prediction model, the set of scalars, and the second set of fluence maps comprises:

calculating a fourth dose using the first dose prediction model and the residual value, and

summing the fourth dose and products of multiplying the set of scalars and the third doses.

16. The method of claim 14 , further comprising accessing third doses comprising differences between doses calculated using the first set of fluence maps and the first dose prediction model and doses calculating using the first set of fluence maps and the second dose prediction model;

wherein calculating the second dose using the first dose prediction model, the set of scalars, and the second set of fluence maps comprises:

calculating a fourth dose using the second fluence map and the first dose prediction model; and

summing the fourth dose and products of multiplying the set of scalers and the third doses.

17. The method of claim 14 , wherein the criterion comprises a threshold value for a magnitude of the residual value.

18. The method of claim 14 , wherein the criterion comprises a threshold value for normalized value of the residual value.

19. The method of claim 14 , wherein the second dose prediction model is more accurate and utilizes more computational resources than the first dose prediction model.

20. The method of claim 14 , further comprising:

selecting a final fluence map for the final radiation treatment plan, wherein the final fluence map is selected from a plurality of fluence maps comprising at least the first set of fluence maps and the second fluence map; and

converting the final fluence map into leaf sequence settings that control movement of a multileaf collimator.