Decision support tool for choosing treatment plans

Patent metadata

Field	Value
Publication number	US-10762167-B2
Application number	US-201314040479-A
Country	US
Kind code	B2
Filing date	Sep 27, 2013
Priority date	Sep 27, 2013
Publication date	Sep 1, 2020
Grant date	Sep 1, 2020

Abstract

Patient data can be used to determine input values to different estimation functions for different treatment types. The estimation functions can each be used to estimate one or more outcome values for the respective treatment. A quality score can be determined using the outcome value(s). A first treatment plan having an optimal quality score can be identified, e.g., by displaying the treatment plans with the quality scores, which may correspond to the outcome values.

First claim

What is claimed is:

1. A method of selecting among a plurality of treatment plan types for treating a tumor of a first patient, the method comprising:
for each of the plurality of treatment plan types:
creating, by a computer system, an estimation function associated with the treatment plan type by:
receiving previous outcome values of one or more types of outcome values for previous patients treated using the treatment plan type,
receiving previous input values of input variables for the previous patients, for each of the one or more types of outcome values:
using the previous input values and the previous outcome values corresponding to the type of outcome value to determine an estimation subfunction of the estimation function;

receiving, by the computer system, information about the first patient;
analyzing, by the computer system, the information to determine a set of input values, the input values including characteristics of the tumor;
for each of the plurality of treatment plan types:
identifying an estimation function associated with the treatment plan type, and
predicting a quality of the treatment plan type by using the set of input values with the identified estimation function to calculate, by the computer system, one or more outcome values, and determining a quality score based on the one or more outcome values, the set of input values including the one or more dose input values;

identifying, by the computer system, a first treatment plan type having an optimal quality score to a user of the computer system, the first treatment plan type including providing radiation to the first patient;
generating, by the computer system, a first treatment plan of the first treatment plan type based on the set of input values and the plurality of outcome values, the first treatment plan including instructions for controlling a treatment head in order to treat the tumor of the first patient; and
providing, by the treatment head coupled with a radiation source, radiation at pre-defined angles and pre-defined doses to specific portions of a treatment area of the first patient according to the first treatment plan, wherein the treatment head is controlled by a control unit according to the instructions in the first treatment plan.

2. The method of claim 1 , further comprising:
comparing, by the computer system, the quality scores of the treatment types to identify the first treatment plan type having the optimal quality score.

3. The method of claim 1 , wherein calculating a first outcome value includes:
predicting a dose distribution based on the set of input values; and
using the predicted dose distribution to determine the first outcome value.

4. The method of claim 1 , wherein using the previous input values and the previous outcome values corresponding to the type of outcome value to determine an estimation subfunction of the estimation function includes:
plotting data points of the previous input values and the previous outcome values corresponding to the type of outcome value; and
using a distance between the data points to compute a mapping function from the input variables to the previous outcome values, wherein the mapping function corresponds to the estimation subfunction of the estimation function.

5. The method of claim 1 , wherein identifying the first treatment plan type includes:
outputting the treatment plan types with the corresponding quality scores.

6. The method of claim 1 , wherein the treatment plan types include chemotherapy and surgery.

7. The method of claim 1 , wherein a plurality of outcome values are calculated from each of at least one of the respective estimation functions corresponding to at least one of the treatment plan types, the method further comprising:
for each of the at least one of the treatment plan types:
using a quality function to determine the quality score from the plurality of outcome values, wherein each outcome value is determined by a respective estimation subfunction.

8. The method of claim 7 , further comprising:
determining the quality function based on criteria provided by the user of the computer system.

9. The method of claim 1 , wherein the quality score is one of the outcome values.

10. The method of claim 9 , wherein identifying the first treatment plan type includes:
sorting the quality scores in response to criteria from the user.

11. A computer product comprising a non-transitory computer readable medium storing a plurality of instructions that when executed control a computer system to select among a plurality of treatment plan types for treating a tumor of a first patient, the instructions comprising:
receiving information about the first patient;
analyzing the information to determine a set of input values, the input values including characteristics of the tumor;
selecting, from a plurality of dose prediction models, a dose prediction model corresponding to a particular treatment plan type;
using the dose prediction model to predict a dose distribution of radiation to be provided to the tumor for the particular treatment plan based on the characteristics of the tumor, the dose distribution specifying a variation of radiation dose within the tumor;
determining one or more dose input values from the predicted dose distribution for the particular treatment plan type;
for each of the plurality of treatment plan types:
identifying an estimation function associated with the treatment plan type, the estimation function generated based on outcomes of other patients using the treatment plan type, and
predicting a quality of the treatment plan type by using the set of input values with the identified estimation function to calculate one or more outcome values, the set of input values including the one or more dose input values for the particular treatment plan type, and by determining a quality score based on the one or more outcome values, the set of input values including the one or more dose input values; and

identifying a first treatment plan type having an optimal quality score to a user of the computer system, the first treatment plan type including providing radiation to the first patient;
generating a first treatment plan of the first treatment plan type based on the set of input values and the plurality of outcome values, the first treatment plan including instructions for controlling a treatment head in order to treat the tumor of the first patient; and
providing, by the treatment head coupled with a radiation source, radiation at pre-defined angles and pre-defined doses to specific portions of a treatment area of the first patient according to the first treatment plan, wherein the treatment head is controlled by a control unit according to the instructions in the first treatment plan.

12. The method of claim 4 , wherein the estimation function for each treatment plan type includes a plurality of estimation subfunctions, wherein each of the one or more outcome values are determined by a respective estimation subfunction, wherein each estimation subfunction corresponds to a multidimensional surface, and wherein each input variable corresponds to a different dimension in the multidimensional surface plot.

13. The method of claim 1 , wherein each outcome value corresponds to either a biological effect on the patient or a property of the treatment.

14. The method of claim 13 , wherein the outcome values include at least one of a probability of treatment success, a quality of life indicator, a probability of organ failure, a treatment time, and a recovery time.

15. The method of claim 1 , wherein analyzing the

Assignees

• Varian Medical Systems Int Ag

Inventors

• Hartman Joona
• Cordero Marcos María Isabel
• Kuusela Esa
• Peltola Jarkko Yrjana
• Nord Janne Ilmari

Classifications

• A61N2005/1041

  using a library of previously administered radiation treatment applied to other patients · CPC title

• A61N5/103 Primary

  Treatment planning systems · CPC title

• G16H50/20 Primary

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

• G06F19/00 Primary

  Physics · mapped topic