Task allocation for use in medical imaging and diagnostics

1. A control method for controlling data processing acquired from medical imaging modalities, by using multiple data processors connected to the multiple medical imaging modalities via a computer network, the control method comprising:

obtaining imaging information for imaging to be performed with an imaging modality of the multiple medical imaging modalities;

obtaining, before the imaging is completed, load information of the multiple data processors;

allocating, based on a graph information generated based on the obtained load information, at least a part of the multiple data processors to processing of data acquired in imaging based on the imaging information; and

performing processing of the acquired data with the allocated data processing resource,

wherein the graph information is a dataflow graph, each node of the dataflow graph specifying a processing step and each edge of the dataflow graph connecting two nodes to specify a route a processing result follows from one processing step to a next processing step, wherein data flowing at each edge of the dataflow graph is not shared among the multiple data processors and each node of the dataflow graph is a computation that can be shared among the multiple data processors.

2. The method of claim 1 , wherein all nodes of the dataflow graph are executed in parallel whose edges have data that is available from prior nodes completing execution.

3. The method of claim 1 , wherein each node from the dataflow graph represents the processing of data acquired in imaging based on the imaging information.

4. The method of claim 3 , wherein a state of the processing of data acquired in imaging based on the imaging information is stored in a memory.

5. The method of claim 4 , wherein a set of all node operands in the dataflow graph are stored in the memory.

6. The method of claim 5 , wherein each medical imaging modality from the multiple medical imaging modalities corresponds to one key-value store.

7. The method of claim 1 , wherein the processing of data acquired in imaging based on the imaging information corresponds to a specific medical imaging modality, the specific medical imaging modality can only access an associated key-value store.

8. The method of claim 1 , wherein the imaging information is expressed within the dataflow graph.

9. The method of claim 1 , further comprising:

assigning priorities to the processing of data acquired in imaging based on the imaging information by a client application; and

executing the processing of data in accordance with available computing resources.

10. The method of claim 1 , wherein the load information is used to determine which data processor from the multiple data processors has a smallest computation cost to process data using the data processor with the smallest computation cost.

11. The method of claim 1 , wherein the multiple medical imaging modalities include any combination from X-ray imaging modality, computed tomography (CT) scan imaging modality, magnetic resonance imaging (MRI) modality, ultrasound imaging modality and PET scan imaging modality.

12. The method of claim 1 , wherein the multiple data processors include at least two of a central processing unit (CPU), graphics processing unit (GPU), hardware compute accelerator, field-programmable gate array (FPGA), Artificial Intelligence (AI) accelerator.

13. The method of claim 1 , wherein the imaging information is obtained from a radiology information system (RIS) or a hospital information system (HIS).

14. The method of claim 1 , wherein each processing step of the dataflow graph is executed in an order that is dependent upon a specification of an imaging modality used for imaging, a type of image reconstruction process to be performed, or a type of image filtering process to be performed.

15. The method of claim 14 , wherein a dataflow name of the dataflow graph enables a processing server to determine the imaging modality used for imaging, the type of image reconstruction process to be performed, or the type of image filtering process to be performed.

16. The method of claim 1 , wherein an imaging modality is designated as a prioritized imaging modality and processing of data associated with the imaging modality that is designated is executed prior to the processing of data associated with an imaging modality that has not been designated.

17. A processing server for controlling data processing acquired from medical imaging modalities, the processing server comprising:

one or more memories storing instructions; and

one or more processors executing the instructions to:

obtain imaging information for imaging to be performed with an imaging modality from multiple medical imaging modalities;

obtain, before the imaging is completed, load information of multiple data processors connected to the multiple medical imaging modalities;

allocate, based on a graph information generated based on the obtained load information, at least a part of the multiple data processors to processing of data acquired in imaging based on the imaging information; and

perform processing of the acquired data with the allocated data processing resource,

wherein the graph information is a dataflow graph, each node of the dataflow graph specifying a processing step and each edge of the dataflow graph connecting two nodes to specify a route a processing result follows from one processing step to a next processing step, wherein data flowing at each edge of the dataflow graph is not shared among the multiple data processors and each node of the dataflow graph is a computation that can be shared among the multiple data processors.

18. A non-transitory computer-readable storage medium storing at least one program that operates to cause one or more processors to execute a control method for controlling data processing acquired from medical imaging modalities, by using multiple data processors connected to the multiple medical imaging modalities via a computer network, the control method comprising:

obtaining imaging information for imaging to be performed with an imaging modality of the multiple medical imaging modalities;

obtaining, before the imaging is completed, load information of the multiple data processors;

allocating, based on a graph information generated based on the obtained load information, at least a part of the multiple data processors to processing of data acquired in imaging based on the imaging information; and

performing processing of the acquired data with the allocated data processing resource,

wherein the graph information is a dataflow graph, each node of the dataflow graph specifying a processing step and each edge of the dataflow graph connecting two nodes to specify a route a processing result follows from one processing step to a next processing step, wherein data flowing at each edge of the dataflow graph is not shared among the multiple data processors and each node of the dataflow graph is a computation that can be shared among the multiple data processors.

19. A method for controlling data processing acquired from medical imaging modalities, by using multiple data processors connected to the multiple medical imaging modalities via a network, the method comprising:

receiving imaging information that is generated by the medical imaging modalities;

compiling a dataflow graph based on the received imaging information, the dataflow graph including a plurality of nodes, wherein each node from the plurality of nodes represents an imaging computation; and

executing the imaging computation for each node of the dataflow graph,

wherein, the execution of each node is allocated to the multiple data processors to generate an image based on the imaging information received from the multiple medical imaging modalities,

wherein each edge of the dataflow graph connects two nodes and data flowing at each edge is not shared among the multiple data processors.