IP Library › Granted Patent US 12,499,549
Granted Patent B2
US 12,499,549 · App. 17/699,837 · Granted Dec 16, 2025

Processing 2-D projection images using a neural network

Inventors: Sailesh Conjeti (Erlangen, DE); Alexander Preuhs (Erlangen, DE)
Assignee: SIEMENS HEALTHINEERS AG
G06T7/11G06T7/0012G06T7/174G06T2207/10116G06T2207/20084
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,499,549
App. No.
17/699,837
Granted
Dec 16, 2025
Kind
B2
Abstract

Techniques are described to infer 2-D segmentations of a region of interest using a neural network algorithm. Techniques are described to train the neural network algorithm. The 2-D segmentations are determined based on multiple 2-D projection images. For example, x-ray images can be used as an input.

Claims (52)

1 . A method, comprising:

obtaining multiple 2-D projection images associated with multiple views of a scene;

determining, using at least one neural network algorithm and for the multiple 2-D projection images, multiple 2-D segmentations of a region of interest included in the scene, the multiple 2-D segmentations being associated with the multiple views; and

based on a predefined registration of the multiple views in a reference frame, determining an inter-view consistency between the multiple 2-D segmentations associated with the multiple views, wherein

the determining the inter-view consistency includes determining a distance between a first reference point of a first one of the multiple 2-D segmentations and a projection of a second reference point of a second one of the multiple 2-D segmentations into a view of the multiple views associated with the first one of the multiple 2-D segmentations,

the first reference point and the second reference point denote a corresponding feature of the region of interest, and

the projection of the second reference point is based on the predefined registration of the multiple views in the reference frame.

2 . The method of claim 1 ,

wherein said determining of the inter-view consistency is based on a fundamental matrix defining the predefined registration.

3 . The method of claim 2 , further comprising:

enforcing the inter-view consistency by adjusting the multiple 2-D segmentations.

4 . The method of claim 1 , further comprising:

enforcing the inter-view consistency by adjusting the multiple 2-D segmentations.

5 . The method of claim 1 ,

wherein the projection of the second reference point into the view associated with the first one of the multiple 2-D segmentations comprises an epipolar line defined in the view associated with the first one of the multiple 2-D segmentations.

6 . The method of claim 1 ,

wherein said determining of the multiple 2-D segmentations is based on a projection matrix defining the predefined registration.

7 . The method of claim 1 , further comprising:

determining a classification of an object defining the region of interest based on the multiple 2-D segmentations.

8 . The method of claim 1 , further comprising:

determining the predefined registration of the multiple views in the reference frame based on at least one of prior-knowledge of an appearance of the region of interest or a configuration of an imaging facility used to acquire the multiple 2-D projection images.

9 . The method of claim 1 , wherein the multiple 2-D projection images are x-ray images.

10 . A method, comprising:

obtaining multiple 2-D projection images associated with multiple views of a scene,

determining, using at least one neural network algorithm and based on the multiple 2-D projection images, a 3-D segmentation of a region of interest included in the scene, and

determining, based on a predefined registration of the multiple views in a reference frame and based on the 3-D segmentation, an inter-view consistency between multiple 2-D segmentations of the region of interest associated with the multiple views,

wherein the determining the inter-view consistency includes determining a distance between a first reference point of a first one of the multiple 2-D segmentations and a projection of a second reference point of a second one of the multiple 2-D segmentations into a view of the multiple views associated with the first one of the multiple 2-D segmentations,

the first reference point and the second reference point denote a corresponding feature of the region of interest, and

the protection of the second reference point is based on the predefined registration of the multiple views in the reference frame.

11 . The method of claim 10 ,

wherein said determining of the multiple 2-D segmentations is based on a projection matrix defining the predefined registration.

12 . The method of claim 10 , further comprising:

determining a classification of an object defining the region of interest based on the multiple 2-D segmentations.

13 . The method of claim 12 , further comprising:

determining the predefined registration of the multiple views in the reference frame based on at least one of prior-knowledge of an appearance of the region of interest or a configuration of an imaging facility used to acquire the multiple 2-D projection images.

14 . The method of claim 10 , further comprising:

determining the predefined registration of the multiple views in the reference frame based on at least one of prior-knowledge of an appearance of the region of interest or a configuration of an imaging facility used to acquire the multiple 2-D projection images.

15 . A device, comprising:

a processor configured to,

obtain multiple 2-D projection images associated with multiple views of a scene,

determine, using at least one neural network algorithm and for the multiple 2-D projection images, multiple 2-D segmentations of a region of interest included in the scene, the multiple 2-D segmentations being associated with the multiple views, and

based on a predefined registration of the multiple views in a reference frame, determine an inter-view consistency between the multiple 2-D segmentations associated with the multiple views, wherein

to determine the inter-view consistency, the processor is configured to determine a distance between a first reference point of a first one of the multiple 2-D segmentations and a projection of a second reference point of a second one of the multiple 2-D segmentations into a view of the multiple views associated with the first one of the multiple 2-D segmentations,

the first reference point and the second reference point is based on the predefined registration of the multiple views in the reference frame.

16 . A device, comprising:

a processor configured to,

obtain multiple 2-D projection images associated with multiple views of a scene,

determine, using at least one neural network algorithm and based on the multiple 2-D projection images, a 3-D segmentation of a region of interest included in the scene, and

determine, based on a predefined registration of the multiple views in a reference frame and based on the 3-D segmentation, an inter-view consistency between multiple 2-D segmentations of the region of interest associated with the multiple views, wherein

to determine the inter-view consistency, the processor is configured to determine a distance between a first reference point of a first one of the multiple 2-D segmentations and a projection of a second reference point of a second one of the multiple 2-D segmentations into a view of the multiple views associated with the first one of the multiple 2-D segmentations,

the first reference point and the second reference point denote a corresponding feature of the region of interest, and

the projection of the second reference point is based on the predefined registration of the multiple views in the reference frame.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 20, 2023
From: SIEMENS HEALTHCARE GMBH
To: SIEMENS HEALTHINEERS AG
Reel/Frame 066267/0346 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 17, 2023
From: CONJETI, SAILESH; PREUHS, ALEXANDER
To: SIEMENS HEALTHCARE GMBH
Reel/Frame 062390/0442 →
Priority Claims (1)
DE 10 2021 202 784.9 · Mar 23, 2021 · national
Continuity (1)
Related Publication 20220309675A1 · Sep 29, 2022
References Cited (37)
US 20050027193A1 · Mitschke · 2005 [cited by examiner]
US 20090021614A1 · Baker · 2009 [cited by examiner]
US 20090074277A1 · Deinzer · 2009 [cited by examiner]
US 20090180678A1 · Kuduvalli · 2009 [cited by examiner]
US 20100021031A1 · Brockway · 2010 [cited by examiner]
US 20120219185A1 · Hu · 2012 [cited by examiner]
US 20130108103A1 · Thueux · 2013 [cited by examiner]
US 20130208948A1 · Berkovich · 2013 [cited by examiner]
US 20130208997A1 · Liu · 2013 [cited by examiner]
US 20140078250A1 · Zhang · 2014 [cited by examiner]
US 20140293028A1 · Nguyen · 2014 [cited by examiner]
US 20150339828A1 · Djelouah · 2015 [cited by examiner]
US 20160093058A1 · Moteki et al. · 2016 [cited by applicant]
US 20160373715A1 · Nguyen · 2016 [cited by examiner]
US 20170148222A1 · Holzer · 2017 [cited by examiner]
US 20200364509A1 · Weinzaepfel · 2020 [cited by examiner]
US 20210103776A1 · Jiang · 2021 [cited by examiner]
US 20210125402A1 · Chui · 2021 [cited by examiner]
US 20210125403A1 · Chui · 2021 [cited by examiner]
US 20220287686A1 · Kruecker · 2022 [cited by examiner]
US 20220309675A1 · Conjeti · 2022 [cited by examiner]
US 20230410364A1 · Tsoref · 2023 [cited by examiner]
US 20240144589A1 · Liu · 2024 [cited by examiner]
CN 104376552A · 2015 [cited by applicant]
CN 111476835A · 2020 [cited by applicant]
Liu, Xinhang, et al. “Unsupervised multi-view object segmentation using radiance field propagation.” Advances in Neural Information Processing Systems 35 (2022): 17730-17743. (Year: 2022). [cited by examiner]
Rana, Pravin Kumar, and Markus Flierl. “Inter-view depth consistency testing in depth difference subspace.” arXiv preprint arXiv: 2301.11752 (2023). (Year: 2023). [cited by examiner]
M. Kurc, O. Stankiewicz and M. Domański, “Depth map inter-view consistency refinement for multiview video,” 2012 Picture Coding Symposium, Krakow, Poland, 2012, pp. 137-140, doi: 10.1109/PCS.2012.6213305. (Year: 2012). [cited by examiner]
S. Jeong, J. Lee, B. Kim, Y. Kim and J. Noh, “Object Segmentation Ensuring Consistency Across Multi-Viewpoint Images,” in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 40, No. 10, pp. 2455-2468, O… [cited by examiner]
Würfl et al. “Estimating the Fundamental Matrix Without Point Correspondences With Application to Transmission Imaging” CVPR 2019. [cited by applicant]
Preuhs, Alexander et al., “Maximum likelihood estimation of head motion using epipolar consistency”, CT-Meeting, 2018. [cited by applicant]
Gundel et al., “Multi-task Learning for Chest X-ray Abnormality Classification on Noisy Labels”; May 2019, IEEE; 2019. [cited by applicant]
Yao, Yuan et al: “Multiview Co-segmentation for Wide Baseline Images using Cross-view Supervision”; In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision; Year: 2020, pp. 1942-1951. [cited by applicant]
Bier et al.;“X-ray-transform Invariant Anatomical Landmark Detection for Pelvic Trauma Surgery”, https://link.springer.com/chapter/10.1007/978-3-030-00937-3_7. [cited by applicant]
Schaffert, R. et al., “Learning an Attention Model for Robust 2-D/3-D Registration Using Point-To-Plane Correspondences”, IEEE Transactions on Medical Imaging. [cited by applicant]
Lin et al., “Focal Loss for Dense Object Detection,” 2017, Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 2980-2988.; 2017. [cited by applicant]
Aichert et al., “Redundancies in X-ray images due to the epipolar geometry for transmission imaging”, The third international conference on image formation in X-ray computed tomography, pp. 333-337. [cited by applicant]
Cited By (1)
US 12,678,115