Intelligent multi-scale medical image landmark detection

1. A method for intelligent multi-scale medical image parsing, the method comprising:

determining, by a processor, a set of actions comprising parametric actions comprising movements in each direction that an artificial agent may select from a set of upward, downward, left, right, forward and backwards movements in a parametric space with respect to a training medical image;

determining, by the processor, scale actions specifying navigation from coarse to fine in a scale space with respect to the training medical image;

establishing, by the processor, a reward system based on applying each action of the set of actions and based on at least one target location of the training medical image; and

learning, by the processor, behavior of the artificial agent to maximize a cumulative future reward value of the reward system by navigating a state space of discrete portions of the training image, wherein the behavior of the artificial agent is a sequence of actions from the set of actions navigating the artificial agent towards the at least one target location of the training medical image until convergence, performing a scale action on the state space, and at a different resolution navigating the artificial agent towards the at least one target location until convergence.

2. The method of claim 1 , further comprising:

applying, by the processor, the learned artificial agent on a test image to automatically parse image content for a landmark location; and

displaying the identified landmark location in the test image.

3. The method of claim 2 , wherein applying the learned artificial agent comprises determining a next action of the artificial agent based on balancing maximization of the cumulative future reward value by actions changing the parametric space and by actions changing the scale space.

4. The method of claim 1 , wherein the at least one target location is an anatomical landmark location, and wherein the target location is defined by a position parameters of the landmark anatomical landmark, and wherein a reward value is indicative of a proximity to the at least one target location.

5. The method of claim 1 , wherein learning the behavior of an artificial agent further comprises:

generating an experience memory database including a predefined number of last evaluated parametric spaces and scale spaces for the training medical image;

sampling the experience memory database; and

updating parameters of the artificial agent based on the experience memory database.

6. The method of claim 1 , wherein learning the behavior of the artificial agent further comprises:

parameterizing the behavior of the artificial agent using a deep neural network; and optimizing the behavior of the artificial agent using an episodic trajectory for the training medical image based on discrete portions of the training medical image via the parametric space and scale space, wherein the episodic trajectory is indicative of a series actions of the set of actions of the artificial agent.

7. The method of claim 1 , wherein the scale actions change the scale space of the state space by increasing or decreasing a resolution with respect to the training medical image.

8. The method of claim 7 , wherein the set of actions further includes an action in which the parametric space is unchanged.

9. A system for training an artificial agent for intelligent multi-scale medical image parsing, the system comprising:

at least one processor; and

at least one memory including computer program code for one or more programs, the at least one memory and the computer program code configured to, with the at least one processor, cause the system to:

receive a plurality of training medical images;

train, based on the plurality of training medical images, an artificial agent to parse a test image to identify a landmark location in the test image, wherein training the artificial agent comprises training a search strategy model to search for the landmark location by parsing the test image by performing a series of actions, the series of actions comprising a set of upward, downward, left, right, forward and backwards movements in a patch of the test image that navigate the artificial agent towards the landmark location, changing a scale of the patch of the test image to parse the test image by searching less than an entire test image, and performing a second series of actions comprising the set of upward, downward, left, right, forward and backwards movements; and

store the trained artificial agent for automatically parsing test image data to identify the landmark location in the test image.

10. The system of claim 9 , wherein searching for the landmark location comprises changing the position of the patch at a first scale, changing the scale from the first scale to a second scale, and changing the position of the patch at the second scale.

11. The system of claim 9 , wherein training the search strategy model comprises encoding parameters of search strategy model in a multilayer data representation.

12. The system of claim 11 , wherein the multilayer data representation is a deep neural network.

13. The system of claim 12 , wherein training the artificial agent comprises using experience memory from previously parsed patches at different scales to solve for parameters of the deep neural network.

14. The system of claim 9 , wherein training the search strategy model comprises maximizing a future reward using a reward system based on reward values for each position and scale change of the patch.

15. A method for intelligent multi-scale landmark identification in a medical image, the method comprising:

parameterizing a patch of the medical image in a trained hierarchical data representation, the hierarchical data representation trained to navigate an artificial agent towards a landmark location by maximizing a future reward of a reward system for each of a plurality of available actions that reposition the patch in an upward, downward, left or right direction in the medical image;

rescaling the patch to a higher or lower resolution;

repeating the training of the hierarchical data representation to navigate the artificial agent towards the landmark location in the rescaled patch;

identifying the landmark location in the rescaled patch of the medical image; and

displaying the medical image with an annotation indicative of the identified landmark location.

16. The method of claim 15 , wherein identifying the landmark location comprises performing a sequence of actions to move a location of the patch toward a location of a target patch and to increase the resolution of the patch, the target patch including the landmark location in the medical image.

17. The method of claim 16 , wherein the sequence of actions comprises a path converging on the landmark location by parsing less than an entire medical image.

18. The method of claim 17 , wherein determining the sequence of actions comprises parameterizing at least one previous patch in the hierarchical data representation after repositioning and rescaling the patch.

19. The method of claim 18 , wherein the hierarchical data representation is a deep neural network.