| CPC A61B 34/10 (2016.02) [A61B 8/4254 (2013.01); A61B 8/463 (2013.01); A61B 8/5261 (2013.01); A61B 17/3403 (2013.01); G06T 5/70 (2024.01); G06T 7/344 (2017.01); G06T 7/38 (2017.01); G06T 15/08 (2013.01); G06T 17/20 (2013.01); G16H 30/20 (2018.01); G16H 30/40 (2018.01); A61B 2017/3411 (2013.01); A61B 2017/3413 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/10028 (2013.01); G06T 2207/10088 (2013.01); G06T 2207/10132 (2013.01); G06T 2207/30096 (2013.01); G06T 2207/30196 (2013.01); G06T 2210/41 (2013.01); G06T 2210/56 (2013.01)] | 17 Claims |

|
1. A method for an ultrasonic puncture guidance planning system based on a multi-modal medical image registration, comprising:
(a) receiving magnetic resonance imaging (MRI) sequence images of a human body, performing three-dimensional rendering on the MRI sequence images and plotting a puncture point;
(b) receiving ultrasound sequence images of the human body, extracting three-dimensional coordinates of four vertices of each frame of the ultrasound sequence images, and reconstructing ultrasound volume data;
(c) performing, by using an Iterative Closest Point (ICP) iterative algorithm, a three-dimensional registration operation on (i) the ultrasound volume data reconstructed in step (b) and (ii) the MRI data rendered in step (a) to obtain a registration transformation coefficient that transforms three-dimensional coordinates corresponding to ultrasound data into three-dimensional coordinates corresponding to MRI data, wherein the registration operation comprises (1) performing triangle-mesh processing and smoothing on the reconstructed ultrasound volume data and down-sampling to obtain a three-dimensional source point cloud of the reconstructed ultrasound volume data, and (2) performing structural pairing on the three-dimensional source point cloud and three-dimensional target point cloud of the MRI-rendered data using a K-Nearest Neighbor (KNN) tree, determining corresponding point pairs, and executing the ICP iterative algorithm to complete registration, thereby obtaining a registration rotation matrix R and a registration translation amount T; and
(d) receiving ultrasound images of the human body acquired by an ultrasound probe, and generating and displaying a puncture planning image based on a system puncture point and a current ultrasound image.
|