Determination of breathing signal from thermal images

1. A method implemented by a computer having one or more processors, comprising:

acquiring image data representing a sequence of training thermal images of at least a part of a surface of a patient's body over time, the sequence covering at least one half breathing cycle and being captured by a thermographic camera;

tracking at least one tracked point in the image data over the sequence of training thermal images to find a trajectory of the at least one tracked point as breathing signal data, wherein the at least one tracked point is a point on the surface of the patient's body; and

defining a gating region in a live thermal image captured by the thermographic camera, wherein the gating region is an area of the live thermal image and wherein at least a part of the trajectory lies in the gating region;

finding a position of the at least one tracked point in the live thermal image; and

generating a gating signal indicating whether or not the at least one tracked point is within the gating region.

2. The method of claim 1 , further comprising the steps of defining a corridor around the trajectory in a live thermal image captured by the thermographic camera, finding the position of the at least one tracked point in the live thermal image and outputting a movement warning signal if the position of the at least one tracked point is outside the corridor.

3. The method of claim 1 , further comprising the steps of determining the speed of the at least one tracked point from at least two consecutive live thermal images captured by the thermographic camera, comparing the determined speed with the speed of the at least one tracked point at a corresponding position of the trajectory and outputting a speed warning if the difference of the speeds is above a predetermined threshold.

4. The method of claim 1 , further comprising a step of a dimension reduction of the trajectory into one dimension.

5. The method of claim 4 , wherein the dimension reduction step includes determining a main axis of the trajectory and projecting the trajectory onto the main axis.

6. The method of claim 1 , wherein the at least one tracked point is selected automatically.

7. The method of claim 6 , wherein the at least one tracked point is selected automatically within a user-defined area of the thermal image.

8. The method of claim 1 , wherein a plurality of tracked points are tracked in the sequence of training thermal images to find trajectories of each of the tracked points.

9. The method of claim 8 , wherein the plurality of tracked points comprises only tracked points with similar main axes of their trajectories.

10. The method of claim 8 , further comprising the steps of defining a gating region for each of the plurality of tracked points in a live thermal image captured by the thermographic camera, finding the positions of each of the tracked points in the live thermal image and generating a gating signal indicating whether or not all or the majority of the tracked points are within their respective gating region.

11. The method of claim 8 , further comprising the steps of determining a main axis of the trajectory of each of the tracked points, projecting the trajectories onto the respective main axis and averaging the projected trajectories to obtain one-dimensional breathing signal data.

12. The method of claim 1 , further comprising the step of correlating the trajectory to a movement of a tumour as described by a tumour movement model.

13. A non-transitory computer readable storage medium comprising stored instructions executable by at least one processor to:

acquire image data representing a sequence of training thermal images of at least a part of the surface of the patient's body over time, the sequence covering at least one half breathing cycle and being captured by a thermographic camera;

track at least one tracked point in the image data over the sequence of training thermal images to find a trajectory of the at least one tracked point as the breathing signal data, wherein the at least one tracked point is a point on the surface of the patient's body; and

define a gating region in a live thermal image captured by the thermographic camera, wherein the gating region is an area of the live thermal image and wherein at least a part of the trajectory lies in the gating region;

find a position of the at least one tracked point in the live thermal image; and

generate a gating signal indicating whether or not the at least one tracked point is within the gating region.

14. A system including memory and one or more processors operable to execute instructions stored in the memory, comprising instructions to:

acquire image data representing a sequence of training thermal images of at least a part of the surface of the patient's body over time, the sequence covering at least one half breathing cycle and being captured by a thermographic camera;

track at least one tracked point in the image data over the sequence of training thermal images to find a trajectory of the at least one tracked point as the breathing signal data, wherein the at least one tracked point is a point on the surface of the patient's body; and

define a gating region in a live thermal image captured by the thermographic camera, wherein the gating region is an area of the live thermal image and wherein at least a part of the trajectory lies in the gating region;

find a position of the at least one tracked point in the live thermal image; and

generate a gating signal indicating whether or not the at least one tracked point is within the gating region.