Estimation device, estimation method, and estimation program

1. An estimation device comprising:

at least one processor,

wherein the processor functions as a trained neural network that derives an estimation result relating to a three-dimensional bone density of a bone part from a simple radiation image acquired by simply imaging a subject including the bone part, or a DXA scanning image acquired by imaging the subject by a DXA method, and

the trained neural network learns using, as teacher data, (i) two radiation images acquired by imaging the subject including the bone part with radiation having different energy distributions or the DXA scanning image, and a two-dimensional bone density of the bone part included in the two radiation images or the DXA scanning image, or (ii) the radiation image or the DXA scanning image of the subject or a bone part image representing the bone part of the subject, the two-dimensional bone density of the bone part included in the radiation image, the DXA scanning image, or the bone part image, and the three-dimensional bone density of the bone part of the subject.

2. The estimation device according to claim 1 ,

wherein the three-dimensional bone density is derived from a three-dimensional image of the subject.

3. The estimation device according to claim 2 ,

wherein the three-dimensional image is a CT image.

4. The estimation device according to claim 3 ,

wherein the three-dimensional bone density is obtained by specifying a bone region in the CT image, deriving an attenuation coefficient of radiation in the bone region, and deriving the three-dimensional bone density based on a bone density at each position in the bone region, which is derived based on the attenuation coefficient of the radiation and a mass attenuation coefficient in the bone region.

5. The estimation device according to claim 1 ,

wherein the two-dimensional bone density is derived from the two radiation images or the DXA scanning image.

6. The estimation device according to claim 5 ,

wherein the two-dimensional bone density is derived based on a body thickness distribution of the subject estimated based on at least one radiation image of the two radiation images or the DXA scanning image, an imaging condition in a case of acquiring the two radiation images or the DXA scanning image, and a pixel value of a bone region in the bone part image obtained by extracting the bone part, which is derived by energy subtraction processing of performing weighting subtraction on the two radiation images or the DXA scanning image.

7. The estimation device according to claim 1 ,

wherein the two-dimensional bone density is derived from the bone part image.

8. The estimation device according to claim 7 ,

wherein the two-dimensional bone density is derived based on a body thickness distribution of the subject estimated based on the radiation image or the DXA scanning image, an imaging condition in a case of acquiring the radiation image or the DXA scanning image, and a pixel value of a bone region in the bone part image.

9. An estimation method comprising:

using a trained neural network that derives an estimation result relating to a three-dimensional bone density of a bone part from a simple radiation image acquired by simply imaging a subject including the bone part, or a DXA scanning image acquired by imaging the subject by a DXA method to derive the estimation result relating to the three-dimensional bone density of the bone part,

wherein the trained neural network learns using, as teacher data, (i) two radiation images acquired by imaging the subject including the bone part with radiation having different energy distributions or the DXA scanning image, and a two-dimensional bone density of the bone part included in the two radiation images or the DXA scanning image, or (ii) the radiation image or the DXA scanning image of the subject or a bone part image representing the bone part of the subject, the two-dimensional bone density of the bone part included in the radiation image, the DXA scanning image, or the bone part image, and the three-dimensional bone density of the bone part of the subject.

10. A non-transitory computer-readable storage medium that stores an estimation program causing a computer to execute a procedure comprising:

using a trained neural network that derives an estimation result relating to a three-dimensional bone density of a bone part from a simple radiation image acquired by simply imaging a subject including the bone part, or a DXA scanning image acquired by imaging the subject by a DXA method to derive the estimation result relating to the three-dimensional bone density of the bone part,

wherein the trained neural network learns using, as teacher data, (i) two radiation images acquired by imaging the subject including the bone part with radiation having different energy distributions or the DXA scanning image, and a two-dimensional bone density of the bone part included in the two radiation images or the DXA scanning image, or (ii) the radiation image or the DXA scanning image of the subject or a bone part image representing the bone part of the subject, the two-dimensional bone density of the bone part included in the radiation image, the DXA scanning image, or the bone part image, and the three-dimensional bone density of the bone part of the subject.