Machine learning prediction of dose volume histogram shapes
A control circuit accesses a plurality of information items that each correspond to a resultant dose volume histogram shape for a corresponding different radiation treatment plan. The control circuit then trains a machine learning model to predict a desired dose volume histogram shape using that plurality of information items as a training corpus.
1 . A method comprising:
accessing at least one clinical goal corresponding to radiation treatment for a patient;
accessing a machine learning model that has been trained with a training corpus comprising a plurality of information items that each correspond to a resultant dose volume histogram shape for a corresponding different radiation treatment plan;
while iteratively optimizing a radiation treatment plan for the patient as a function of the at least one clinical goal, predicting information regarding a reference dose volume histogram shape using the machine learning model and using the predicted information to influence optimization of the radiation treatment plan, wherein predicting the information regarding a reference dose volume histogram shape occurs a plurality of times while iteratively optimizing the radiation treatment plan for the patient.
2 . The method of claim 1 wherein at least a substantial majority of the different radiation treatment plans each correspond to a radiation treatment plan for a same patient treatment volume.
3 . The method of claim 2 wherein all of the different radiation treatment plans each correspond to a radiation treatment plan for a same patient treatment volume.
4 . The method of claim 1 wherein the machine learning model is configured to predict confidence intervals for dose volume histogram variances based, at least in part, on at least one of given clinical metric values and patient anatomy.
5 . The method of claim 1 wherein predicting the information regarding a reference dose volume histogram shape comprises, at least in part, generating estimation curves.
6 . The method of claim 1 wherein predicting the information regarding a reference dose volume histogram shape comprises, at least in part, generating information regarding at least one of a predicted dose volume histogram curve mean and a predicted dose volume histogram variance.
7 . The method of claim 1 wherein predicting the information regarding a reference dose volume histogram shape comprises, at least in part, generating information regarding at least both of a predicted dose volume histogram curve mean and a predicted dose volume histogram variance.
8 . The method of claim 1 wherein using the predicted information to influence optimization of the radiation treatment plan comprises, at least in part, using the predicted information to create a cost function term.
9 . The method of claim 8 wherein using the predicted information to create a cost function term comprises using the predicted information to create a cost function term for each structure having an associated clinical goal.
10 . The method of claim 1 further comprising:
using a corresponding optimized radiation treatment plan to administer therapeutic radiation to the patient.
11 . An apparatus comprising:
a control circuit configured to:
access at least one clinical goal corresponding to radiation treatment for a patient;
access a machine learning model that has been trained with a training corpus comprising a plurality of information items that each correspond to a resultant dose volume histogram shape for a corresponding different radiation treatment plan;
while iteratively optimizing a radiation treatment plan for the patient as a function of the at least one clinical goal, predict information regarding a reference dose volume histogram shape using the machine learning model and use the predicted information to influence optimization of the radiation treatment plan, wherein predicting the information regarding a reference dose volume histogram shape occurs a plurality of times while iteratively optimizing the radiation treatment plan for the patient.
12 . The apparatus of claim 11 wherein the control circuit is configured to predict the information regarding a reference dose volume histogram shape by, at least in part, generating information regarding at least one of a predicted dose volume histogram curve mean and a predicted dose volume histogram variance.
13 . The apparatus of claim 11 wherein the control circuit is configured to use the predicted information to influence optimization of the radiation treatment plan by, at least in part, using the predicted information to create a cost function term.
14 . The apparatus of claim 13 wherein the control circuit is configured to use the predicted information to create a cost function term by using the predicted information to create a cost function term for each structure having an associated clinical goal.