US 12,213,836 B2
Methods and systems for segmentation and rendering of inverted data
Paul Sheeran, Woodinville, WA (US); Thanasis Loupas, Kirkland, WA (US); and Charles Tremblay-Darveau, Seattle, WA (US)
Assigned to KONINKLIJKE PHILIPS N.V., Eindhoven (NL)
Appl. No. 17/431,491
Filed by KONINKLIJKE PHILIPS N.V., Eindhoven (NL)
PCT Filed Feb. 21, 2020, PCT No. PCT/EP2020/054631
§ 371(c)(1), (2) Date Aug. 17, 2021,
PCT Pub. No. WO2020/169805, PCT Pub. Date Aug. 27, 2020.
Claims priority of provisional application 62/808,572, filed on Feb. 21, 2019.
Prior Publication US 2022/0133278 A1, May 5, 2022
Int. Cl. A61B 8/08 (2006.01); G06N 7/02 (2006.01); G06T 15/08 (2011.01)
CPC A61B 8/483 (2013.01) [A61B 8/0866 (2013.01); A61B 8/5269 (2013.01); G06N 7/02 (2013.01); G06T 15/08 (2013.01)] 13 Claims
OG exemplary drawing
 
1. A method comprising:
sorting data points of a three dimensional (3D) dataset acquired by an ultrasound imaging system into one of a first set or a second set, wherein the 3D dataset defines a first volume;
generating a mask corresponding to the 3D dataset, wherein data points of the mask corresponding to data points of the 3D dataset sorted into the first set are set to a first value and data points of the mask corresponding to data points of the 3D dataset sorted into the second set are set to a second value, wherein an outer boundary of the data points of the mask set to the first value defines a second volume within the first volume;
applying the mask to the 3D dataset such that data points of the 3D dataset corresponding to data points outside the second volume are discarded to generate a subset of the 3D dataset; and
generating an inverted render from the subset of the 3D dataset, wherein generating the inverted render comprises rendering data points representing higher intensity signals with darker pixels than pixels used to render data points representing lower intensity signals,
further comprising filtering the 3D dataset to increase a difference between signal values and noise values prior to sorting the data points of the 3D dataset.