Medical image processing apparatus

1 . A medical image processing apparatus, comprising:

processing circuitry configured to:

acquire a medical image for setting identification information for identifying a plurality of regions in the medical image;

determine a plurality of seed points in the medical image;

select one of the plurality of seed points on a shortest path from a first node, the shortest path being a path in which a sum of energies between pixels in the medical image is minimized, and the first node being a pixel of the medical image for which the identification information is not set; and

allocate the identification information to the first node,

wherein the processing circuitry is further configured to generate a graph structure including a plurality of first nodes and connecting the plurality of seed points and the plurality of first nodes,

the graph structure further includes a second node provided outside the medical image and connected to the plurality of seed points, and

the processing circuitry is further configured to select the shortest path from the second node to the first node, thereby selecting one of the plurality of seed points on the shortest path from the first node.

2 . The medical image processing apparatus of claim 1 , wherein the plurality of regions includes an anatomical structure of a human body.

3 . The medical image processing apparatus of claim 1 ,

wherein the graph structure includes a plurality of first edges connecting the plurality of seed points and the plurality of first nodes, and

wherein the processing circuitry is further configured to apply energy to the plurality of first edges based on a difference between the pixels in the medical image.

4 . The medical image processing apparatus of claim 3 , wherein the processing circuitry is further configured to apply the energy to the plurality of first edges based on a difference in pixel values between the pixels in the medical image.

5 . The medical image processing apparatus of claim 4 , wherein the processing circuitry is further configured to apply the energy to the plurality of first edges based on a difference in existence probability between the pixels in the medical image, the existence probability being a probability that the pixel belongs to one of the plurality of regions.

6 . The medical image processing apparatus of claim 1 ,

wherein the graph structure includes a plurality of second edges connecting the plurality of seed points with the second node, and

wherein the processing circuitry is further configured to apply a same energy to the plurality of second edges.

7 . The medical image processing apparatus of claim 1 , wherein the processing circuitry is further configured to:

perform, on the medical image, filter processing of emphasizing an outline of each of the plurality of regions in the medical image; and

generate the graph structure with, as a boundary, the outline of each of the plurality of regions emphasized by the filter processing.

8 . The medical image processing apparatus of claim 4 , wherein the processing circuitry is further configured to select the shortest path from the second node to the first node so that a sum of energies applied to the first edges from the second node to the first node is minimized, thereby selecting one of the plurality of seed points on the shortest path from the first node.

9 . The medical image processing apparatus of claim 1 , wherein the processing circuitry is further configured to select one of the plurality of seed points on the shortest path from the first node with a Dijkstra method.

10 . The medical image processing apparatus of claim 1 , wherein the processing circuitry is further configured to receive, from an operator, an input operation of determining the plurality of seed points in the medical image and determine the plurality of seed points in the medical image according to the input operation.

11 . The medical image processing apparatus of claim 1 , wherein the processing circuitry is further configured to determine the plurality of seed points in the medical image with a learned model machine-learned to determine the plurality of seed points in the medical image by receiving an input of the medical image.

12 . The medical image processing apparatus of claim 1 , wherein the processing circuitry is further configured to determine the plurality of seed points in the medical image based on a CT value of the medical image.

13 . The medical image processing apparatus of claim 1 , wherein the processing circuitry is further configured to extract a specific region in the medical image and determine the plurality of seed points in the medical image based on the specific region.