Three-dimensional ultrasound reconstruction with confidence information
For simpler scanning in 3D ultrasound imaging, free hand 3D ultrasound imaging is used. Various registration techniques, such as filtering of position signals based on density of scanning and/or counteracting depth deformation due to pressure, may be used for aligning frames relative to each other in 3D space. To deal with curvature of the skin surface resulting in different frames of data representing the same locations, the frames are grouped by orientation. Data representing the same locations in frames of the same or similar orientation are combined. Frame groups of different orientations but of a same sweep of the transducer along the skin are combined into a volume. Volumes from different sweeps are combined. For combining data representing a same location, the values are weighted based on the likelihoods of ultrasound values representing different locations being accurate or correct.
1. A method for three-dimensional ultrasound imaging, the method comprising:
acquiring, with an ultrasound system including a transducer, first and second ultrasound data representing a same volume of a patient, at least some of the first ultrasound data representing the same locations as at least some of the second ultrasound data;
assigning, by a processor, confidence values as a function of location within the volume to the first and second ultrasound data;
combining, by the processor and for each of the same locations, the first and second ultrasound data representing each of the same locations within the volume, the combining being a function of the confidence values for the locations; and
generating an image on a display using results of the combining, the image representing the volume of the patient.
2. The method of claim 1 wherein acquiring comprises acquiring with free hand motion of the transducer, the first and second ultrasound data being from first and second sweeps of the transducer in different directions.
3. The method of claim 1 wherein acquiring comprises acquiring with the transducer along a curved skin surface, the first and second ultrasound data being for frames along planes at different angles relative to the volume due to the curved skin surface.
4. The method of claim 1 wherein assigning comprises assigning the confidence values as a function of scan format and depth.
5. The method of claim 1 wherein assigning comprises assigning the confidence values as a function of structure detected in the volume.
6. The method of claim 1 further comprising transforming the locations from planes into three-dimensional space as a function of signals from a sensor on the transducer.
7. The method of claim 6 further comprising low pass filtering the signals from the sensor, a kernel size of the low pass filtering being a function of a density of the planes.
8. The method of claim 1 further comprising spatially registering frames of the first ultrasound data, the second ultrasound data, or the first and second ultrasound data.
9. The method of claim 8 wherein spatially registering comprises spatially registering patches and generating a depth-based deformation field as a function of results of the spatially registering of the patches.
10. The method of claim 1 wherein combining comprises interpolating with weights that are a function of the confidence values and distance to voxel centers.
11. The method of claim 1 wherein combining comprises weighting the first ultrasound data with confidence values for the first ultrasound data, weighting the second ultrasound data with confidence values for the second ultrasound data, summing results of the weighting for each location, and dividing the summed results by a sum of the confidence values for the first and second ultrasound data.
12. The method of claim 1 further comprising registering the first ultrasound data with the second ultrasound data in three-dimensional space, and wherein combining comprises combining as a function of the registering in the three-dimensional space.
13. The method of claim 1 further comprising selecting a maximum confidence for each location and generating a confidence image as a function of the maximum confidences.
14. The method of claim 1 further comprising:
determining corners represented by the frames of the first ultrasound data relative to the volume, the frames representing planes of the volume; and
identifying bricks of sub-volumes of the first ultrasound data along a scan line based on the corners;
wherein combining comprises combining using the identified bricks.
15. The method of claim 1 where the first and second ultrasound data comprise tissue strain, power Doppler, contrast agent, tissue harmonics, elastic modulus, relaxation time or other tissue property.
16. The method of claim 1 further comprising:
providing a user interface to interact with the display to perform quantitative measurements of volume, area, length, average intensity, intensity profile;
reformatting a curved surface or path to straight; and
calculating a quantity as a function of the reformatted curved surface or path in response to the user interface.