Pulmonary function identifying and treating method
A pulmonary function identifying and treating method includes: obtaining a first image, having first image elements, and a second image, having second image elements, respectively corresponding to a first state and a second state of a lung; extracting first feature points of the first image and second feature points of the second image; registering the first image with the second image using a boundary point set registration method and an inner tissue registration method according to the first feature points and the second feature points, so that the first image elements correspond to the second image elements and tissue units of the lung; determining functional index values representative of the tissue units of the lung using a ventilation function quantification method according to the first image elements and the second image elements corresponding to the first image elements; and treating the lung according to the functional index values.
1. A pulmonary function identifying and treating method, comprising, in order, steps of:
obtaining, by a scanner, a first image of a patient, having first image elements, and a second image of the patient, having second image elements, respectively corresponding to a first state and a second state of a lung of the patient;
receiving, by a computer, the first image and the second image and extracting, by the computer, first feature points of the first image and second feature points of the second image, wherein a count of the first feature points is different from a count of the second feature points, the first feature points are located around and inside the first image, and the second feature points are located around and inside the second image;
registering, by the computer, the first image with the second image using a boundary point set registration method and an inner tissue registration method according to the first feature points and the second feature points, so that the first image elements correspond to the second image elements and tissue units of the lung;
determining, by the computer, functional index values representative of the tissue units of the lung using a ventilation function quantification method according to the first image elements and the second image elements corresponding to the first image elements;
performing, by the computer, radiation therapy planning according to the functional index values to obtain radiation therapy planning data;
receiving, by a radiation therapy apparatus, the radiation therapy planning data; and
treating the patient by the radiation therapy apparatus according to the radiation therapy planning data, wherein:
the boundary point set registration method is a coherent point drift (CPD) point set registration method, and the inner tissue registration method is a Demons image registration method utilizing gradient information of image intensities of the first image and the second image to register the first image and the second image with each other; and
a displacement vector field is obtained using the CPD point set registration method, and the displacement vector field is configured to an initial displacement vector field used in the Demons image registration method.
2. The pulmonary function identifying and treating method according to claim 1 , wherein:
the first image is a first three-dimensional computed tomography (3D-CT) image, and the second image is a second 3D-CT image; or
the first image is a first two-dimensional computed tomography (2D-CT) image, and the second image is a second 2D-CT image.
3. The pulmonary function identifying and treating method according to claim 2 , further comprising a step of:
obtaining the first 3D-CT image and the second 3D-CT image according to a four-dimensional computed tomography (4D-CT) image of the lung, wherein the first 3D-CT image corresponds to a maximum inhale phase of the lung, and the second 3D-CT image corresponds to a maximum exhale phase of the lung.
4. The pulmonary function identifying and treating method according to claim 2 , wherein in the step of registering the first 3D-CT image with the second 3D-CT image, a first boundary of the lung in the first 3D-CT image and a second boundary of the lung in the second 3D-CT image are firstly determined according to the first feature points and the second feature points using the boundary point set registration method, and then registering of inner tissues of the lung is performed according to the first boundary, the second boundary and relationships between Hounsfield unit (HU) values of the first image elements and the second image elements using the inner tissue registration method.
5. The pulmonary function identifying and treating method according to claim 2 , further comprising a step of: classifying the tissue units into multiple levels of regions according to the functional index values.
6. The pulmonary function identifying and treating method according to claim 5 , further comprising: obtaining ventilation functions of the tissue units, based on the ventilation function quantification method, being functional index values representative of the tissue units, wherein the ventilation functions of the tissue units are classified into at least two levels according to a mean and a standard deviation of ventilation metrics to obtain the levels of regions.
7. The pulmonary function identifying and treating method according to claim 1 , wherein the tissue units are set as corresponding to the second image elements in a one-to-one manner.
8. The pulmonary function identifying and treating method according to claim 2 , wherein boundary tracking technology is adopted to obtain boundaries of the lung in multiple horizontal sections of the first 3D-CT image and the second 3D-CT image, and the boundaries are stacked together to obtain the first feature points and the second feature points.
9. The pulmonary function identifying and treating method according to claim 2 , wherein boundary tracking technology is adopted to generate a 3D boundary to obtain the first feature points and the second feature points according to the first 3D-CT image and the second 3D-CT image.
10. The pulmonary function identifying and treating method according to claim 2 , wherein after the CPD point set registration method is performed with the first feature points and the second feature points serving as boundary feature points to obtain a CPD result comprising a changing vector field of the boundary feature points, the Demons image registration method is performed according to the CPD result in conjunction with multiple Hounsfield Unit (HU) values of HU intensity distributions of the first 3D-CT image and the second 3D-CT image, wherein comparisons of the HU values are performed to obtain inner flow field deformations between the first state and the second state and to obtain two inner deformation vector fields, according to which corresponding relationships between the first image elements and the second image elements are established.
11. The pulmonary function identifying and treating method according to claim 1 , wherein the first image and the second image are magnetic resonance images obtained using magnetic resonance imaging (MRI) examination, or computed tomography (CT) images obtained using dual energy computed tomography.
12. The pulmonary function identifying and treating method according to claim 1 , wherein the Demons image registration method is adopted to constrain obtained displacements by elastic regularization.