Selection of magnetic resonance imaging acquisitions for radiation therapy planning

1 . A medical system comprising:

a memory configured to store machine executable instructions, an analysis algorithm, an assessment algorithm, and a recommender algorithm; wherein the analysis algorithm is configured to output a radiotherapy planning map in response to receiving one or more planning magnetic resonance images descriptive of a region of interest of a subject;

wherein the assessment algorithm is configured to output at least one metric map descriptive of the radiotherapy planning map within the region of interest in response to receiving the output of the analysis algorithm;

wherein the recommender algorithm is configured to output either a pulse sequence selection or a stop acquisition command in response to receiving the one or more planning magnetic resonance images, the output of the analysis algorithm, and the output of the assessment algorithm as input; wherein the pulse sequence selection is a planning magnetic resonance imaging pulse sequence selection and a selection or ranking of a predetermined set of radiation therapy planning pulse sequence commands; and

a computational system, wherein execution of the machine executable instructions causes the computational system to:

receive the one or more planning magnetic resonance images;

receive the radio therapy planning map from the analysis algorithm as output in response to inputting the one or more planning magnetic resonance images into the analysis algorithm;

receive the at least one metric map as output from the assessment algorithm in response to inputting the output of the radio therapy planning map into the assessment algorithm; and

receive the pulse sequence selection or the stop acquisition command as output from the recommender algorithm in response to inputting the one or more planning magnetic resonance images, the output of the analysis algorithm, and the output of the assessment algorithm.

2 . The medical system of claim 1 , wherein the recommender algorithm is further configured to receive at least one additional tomographic medical image descriptive of the region of interest of the subject as input, wherein execution of the machine executable instructions causes the computational system to additionally input the at least one additional tomographic medical image into the recommender algorithm.

3 . The medical system of claim 1 , wherein the assessment algorithm is further configured to receive the one or more planning magnetic resonance images and/or receive at least one further tomographic medical image descriptive of the region of interest of the subject as input, wherein execution of the machine executable instructions further causes the computational system to additionally input the one or more planning magnetic resonance images and/or the at least one further tomographic medical image into the assessment algorithm.

4 . The medical system of claim 1 , wherein the analysis algorithm is further configured to receive at least one supplementary tomographic medical image descriptive of the region of interest of the subject as input, wherein execution of the machine executable instructions causes the computational system to additionally input the at least one supplementary tomographic medical image into the analysis algorithm.

5 . The medical system of claim 1 wherein the analysis algorithm is configured to receive up to a predetermined maximum number of planning magnetic resonance images, wherein the analysis algorithm is configured to receive the one or more planning magnetic resonance images using any one the following algorithm architecture:

wherein the input of the analysis algorithm comprises a separate input channel for each of the planning magnetic resonance images; and

wherein the analysis algorithm comprises an image processing algorithm configured for receiving each of the predetermined maximum number of planning magnetic resonance images individually and in response each outputting an intermediate radiotherapy planning map, wherein the analysis algorithm further comprises an image merging algorithm configured for receiving the intermediate radiotherapy planning map of each image processing algorithm and in response outputting the radiotherapy planning map.

6 . The medical system of claim 1 , wherein the analysis algorithm and the assessment algorithm are implemented as a single algorithm.

7 . The medical system of claim 1 , wherein the radiotherapy planning map comprises any one of the following: an organ segmentation, an identification of abnormal anatomical structures, a tumor segmentation, a pseudo computed tomography image, an electron density map, a radiotherapy dose plan, and combinations thereof.

8 . The medical system of claim 1 , wherein the analysis algorithm is any one of the following: a region growing algorithm, a deformable shape model, a trained neural network, a trained random forest, and a registration algorithm configured to register the one or more planning magnetic resonance images to an anatomical atlas.

9 . The medical system of claim 1 , wherein the assessment algorithm is any one of the following: a Monte-Carlo simulation configured for predicting the stability of the radiotherapy planning map, a model configured to measure a correlation in comparison with a saliency map, a principal components analysis model, an algorithm configured to measure a distance to an average model, a fitness to a statistical appearance model such as a texture analysis or a reconstruction from an archived model.

10 . The medical system of claim 1 , wherein the recommender algorithm is any one of the following: a decision tree algorithm, a trained recommender neural network, and a trained random forest algorithm.

11 . The medical system of claim 1 , wherein the medical system further comprises a magnetic resonance imaging system, wherein the memory further contains the predetermined set of radiation therapy planning pulse sequence commands, wherein the predetermined set of radiation therapy planning pulse sequence commands comprises initial pulse sequence commands, wherein execution of the machine executable instructions further causes the computational system to:

acquire initial k-space data by controlling the magnetic resonance imaging system with the initial pulse sequence commands;

reconstruct an initial planning magnetic resonance image from the initial k-space data, wherein the initial planning magnetic resonance image is added to the one or more planning magnetic resonance images;

wherein execution of the machine executable instructions further causes the computational system to repeatedly perform the following steps until receiving the stop command is received:

receive the radio therapy planning map from the analysis algorithm as output in response to inputting the one or more planning magnetic resonance images into the analysis algorithm;

receive the at least one metric map as output from the assessment algorithm in response to inputting the output of the radio therapy planning map into the assessment algorithm;

receive the pulse sequence selection or the stop acquisition command as output from the recommender algorithm in response to inputting the one or more planning magnetic resonance images, the output of the analysis algorithm, and the output of the assessment algorithm;

select subsequent pulse sequence commands from the predetermined set of radiation therapy planning pulse sequence commands using the pulse sequence selection;

acquire subsequent k-space data by controlling the magnetic resonance imaging system with the subsequent pulse sequence commands; and

reconstruct a subsequent planning magnetic resonance image from the subsequent k-space data, wherein the subsequent planning magnetic resonance image is added to the one or more planning magnetic resonance images.

12 . The medical system of claim 11 , wherein the memory further contains a radiotherapy planning module configured to generate radio therapy control commands in response to receiving the radiotherapy planning map, wherein execution of the machine executable instructions further cause the computational system to receive the radiotherapy control commands from the radiotherapy planning module in response to inputting the radiotherapy planning map into the radiotherapy planning module.

13 . The medical system of claim 12 , wherein the radiotherapy planning map is further configured to receive a planning computed tomography image descriptive of the subject and the radiotherapy planning map as input, wherein the planning computed tomography image is a measured computed tomography image or a pseudo computed tomography image, wherein execution of the machine executable instructions further causes the computational system to receive the radiotherapy control commands from the radiotherapy planning module in response to inputting the radiotherapy planning map and the planning computed tomography image into the radiotherapy planning module.

14 . A method of operating a medical system, wherein the method comprises:

receiving one or more planning magnetic resonance images descriptive of a region of interest of a subject;

receiving a radio therapy planning map from an analysis algorithm as output in response to inputting the one or more planning magnetic resonance images into the analysis algorithm, wherein the analysis algorithm is configured to output the radiotherapy planning map in response to receiving the one or more planning magnetic resonance images descriptive of the region of interest of the subject;

receiving at least one metric map as output from an assessment algorithm in response to inputting the output of the radio therapy planning map into the assessment algorithm, wherein the assessment algorithm is configured to output the at least one metric map descriptive of the radiotherapy planning map within the region of interest in response to receiving the output of the analysis algorithm; and

receiving a pulse sequence selection or a stop acquisition command as output from a recommender algorithm in response to inputting the one or more planning magnetic resonance images, the output of the analysis algorithm, and the output of the assessment algorithm; wherein the recommender algorithm is configured to output either the pulse sequence selection or the stop acquisition command in response to receiving the one or more planning magnetic resonance images, the output of the analysis algorithm, and the output of the assessment algorithm as input; wherein the pulse sequence selection is a selection or ranking of a predetermined set of radiation therapy planning pulse sequence commands.

15 . A computer program comprising machine executable instructions stored on a non-transitory computer readable medium, the computer program including: an analysis algorithm, an assessment algorithm, and a recommender algorithm; wherein the analysis algorithm is configured to output a radiotherapy planning map in response to receiving one or more planning magnetic resonance images descriptive of a region of interest of a subject; wherein the assessment algorithm is configured to output at least one metric map descriptive of the radiotherapy planning map within the region of interest in response to receiving the output of the analysis algorithm; wherein the recommender algorithm is configured to output either a pulse sequence selection or a stop acquisition command in response to receiving the one or more planning magnetic resonance images, the output of the analysis algorithm, and the output of the assessment algorithm as input; wherein the pulse sequence selection is a selection or ranking of a predetermined set of radiation therapy planning pulse sequence commands; wherein execution of the machine executable instructions causes the computational system to:

receive the one or more planning magnetic resonance images;

receive the radio therapy planning map from the analysis algorithm as output in response to inputting the one or more planning magnetic resonance images into the analysis algorithm;

receive the at least one metric map as output from the assessment algorithm in response to inputting the output of the radio therapy planning map into the assessment algorithm; and

receive the pulse sequence selection or the stop acquisition command as output from the recommender algorithm in response to inputting the one or more planning magnetic resonance images, the output of the analysis algorithm, and the output of the assessment algorithm.