CN-115661278-B - Construction method, device and equipment for expanding image sequence

CN115661278BCN 115661278 BCN115661278 BCN 115661278BCN-115661278-B

Abstract

The application discloses a construction method, a construction device and construction equipment for an extended image sequence, relates to the technical field of image sequence reconstruction, and mainly aims to solve the problem that the authenticity of forward projection is poor due to the fact that the authenticity of an existing constructed simulated object extended image sequence is low, so that the difference between a simulated object and an actual projection object is large. The method comprises the steps of obtaining tomographic data of a target projection object, carrying out tomographic image reconstruction processing on the tomographic data to generate a tomographic reconstruction image sequence of the target projection object, wherein the tomographic reconstruction image sequence of the target projection object is used for representing a maximum image sequence reconstructed under the current reconstruction condition, and constructing an expansion image sequence of a simulated projection object matched with the target projection object based on the tomographic reconstruction image sequence of the target projection object, wherein the expansion image sequence is used for representing an expansion image sequence of the simulated projection object which is not reconstructed under the current reconstruction condition and needs to be constructed virtually.

Inventors

YUAN CHAO
TONG LIXIA

Assignees

东软医疗系统股份有限公司

Dates

Publication Date: 20260508
Application Date: 20220916

Claims (6)

1. A method of constructing an extended image sequence, comprising: acquiring tomographic data of a target projection object; Performing tomographic image reconstruction processing on the tomographic data to generate a tomographic reconstructed image sequence of the target projection object, wherein the tomographic reconstructed image sequence of the target projection object is used for representing a maximum image sequence reconstructed under the current reconstruction condition; Constructing an extended image sequence of a simulated projection object matched with the target projection object based on the tomographic reconstructed image sequence of the target projection object, wherein the extended image sequence is used for representing the extended image sequence of the simulated projection object which is not reconstructed under the current reconstruction condition and needs to be constructed virtually; The method comprises the steps of constructing an extended image sequence of a simulated projection object matched with the target projection object based on a tomographic reconstructed image sequence of the target projection object, wherein the extended image sequence comprises any one of the following modes: The method comprises the steps of obtaining an equivalent phantom of the target projection object, wherein the equivalent phantom comprises a tomographic scanning sequence and a position corresponding to an expansion sequence, obtaining an equivalent radius of the position corresponding to the tomographic scanning reconstruction image sequence in the equivalent phantom based on the tomographic scanning reconstruction image sequence of the target projection object, determining the expansion radius corresponding to a simulated projection object of each image in the expansion image sequence based on the equivalent phantom and the equivalent radius, and performing polynomial fitting or neural network model processing on the expansion radius and the tomographic scanning reconstruction image sequence of the target projection object to generate an expansion image sequence; the second mode is that flat image data of the target projection object is obtained, a compression mapping model is built based on a tomographic reconstruction image sequence of the target projection object and the flat image data of the target projection object, reconstruction processing is carried out based on the compression mapping model and the flat image data of the target projection object, and an expansion image sequence is obtained; The third mode comprises the steps of obtaining flat image data of the target projection object, obtaining an equivalent phantom of the target projection object, wherein the equivalent phantom comprises a fault scanning sequence and a position corresponding to an expansion sequence, obtaining a complete sequence of the equivalent phantom through polynomial fitting or neural network model processing on a fault scanning reconstruction image sequence of the target projection object and the flat image data of the target projection object, and extracting the expansion image sequence from the complete sequence of the equivalent phantom.
2. The method as recited in claim 1, further comprising: and correcting the extended image sequence based on the tomographic reconstructed image sequence of the target projection object to obtain a corrected extended image sequence, wherein the correction comprises size and/or definition adjustment.
3. The method of claim 1, wherein after constructing the extended image sequence of the simulated projection object that matches the target projection object based on the tomographic reconstructed image sequence of the target projection object, the method further comprises: and combining the extended image sequence and the tomographic reconstruction image sequence of the target projection object to generate a complete image sequence of the simulated projection object.
4. A construction apparatus for expanding an image sequence, comprising: the first acquisition module is used for acquiring tomographic data of the target projection object; The reconstruction module is used for carrying out tomographic image reconstruction processing on the tomographic data to generate a tomographic reconstructed image sequence of the target projection object, wherein the tomographic reconstructed image sequence of the target projection object is used for representing a maximum image sequence reconstructed under the current reconstruction condition; the construction module is used for constructing an extended image sequence of the simulated projection object matched with the target projection object based on the tomographic reconstructed image sequence of the target projection object, and the extended image sequence is used for representing the extended image sequence of the simulated projection object which is not reconstructed under the current reconstruction condition and needs to be constructed virtually; The method comprises the steps of constructing an extended image sequence of a simulated projection object matched with the target projection object based on a tomographic reconstructed image sequence of the target projection object, wherein the extended image sequence comprises any one of the following modes: The method comprises the steps of obtaining an equivalent phantom of the target projection object, wherein the equivalent phantom comprises a tomographic scanning sequence and a position corresponding to an expansion sequence, obtaining an equivalent radius of the position corresponding to the tomographic scanning reconstruction image sequence in the equivalent phantom based on the tomographic scanning reconstruction image sequence of the target projection object, determining the expansion radius corresponding to a simulated projection object of each image in the expansion image sequence based on the equivalent phantom and the equivalent radius, and performing polynomial fitting or neural network model processing on the expansion radius and the tomographic scanning reconstruction image sequence of the target projection object to generate an expansion image sequence; the second mode is that flat image data of the target projection object is obtained, a compression mapping model is built based on a tomographic reconstruction image sequence of the target projection object and the flat image data of the target projection object, reconstruction processing is carried out based on the compression mapping model and the flat image data of the target projection object, and an expansion image sequence is obtained; The third mode comprises the steps of obtaining flat image data of the target projection object, obtaining an equivalent phantom of the target projection object, wherein the equivalent phantom comprises a fault scanning sequence and a position corresponding to an expansion sequence, obtaining a complete sequence of the equivalent phantom through polynomial fitting or neural network model processing on a fault scanning reconstruction image sequence of the target projection object and the flat image data of the target projection object, and extracting the expansion image sequence from the complete sequence of the equivalent phantom.
5. A storage medium having stored therein at least one executable instruction for causing a processor to perform operations corresponding to the method of constructing an extended image sequence according to any one of claims 1-3.
6. An electronic device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus; the memory is configured to store at least one executable instruction, where the executable instruction causes the processor to perform operations corresponding to the method for constructing an extended image sequence according to any one of claims 1-3.

Description

Construction method, device and equipment for expanding image sequence Technical Field The present application relates to the field of image sequence reconstruction technologies, and in particular, to a method, an apparatus, and a device for constructing an extended image sequence. Background In the current medical diagnosis technology, CT (Computed Tomography), namely the electronic computer tomography technology, has the characteristics of short scanning time, clear images and the like, and is widely used in disease examination and health physical examination. In the iterative reconstruction algorithm of CT, the simulated forward projection data is critical to the algorithm, and directly determines the CT image quality, so the simulated object data for forward projection directly affects the projection result, and in order to obtain the simulated object data close to the real projection object, the simulated object data close to the real projection object needs to be constructed. The model of the simulated object data is derived from the CT reconstructed image, but sometimes the CT reconstructed image cannot completely construct the complete simulated object data, which results in a large difference between the simulated object and the actual projected object, and thus affects the authenticity of the forward projection. Disclosure of Invention In view of this, the present application provides a method, apparatus and device for constructing an extended image sequence, which aims to solve the problem that the reality of forward projection is poor due to the fact that the reality of the extended image sequence of the existing constructed simulated object is low, so that the difference between the simulated object and the real projection object is large. According to one aspect of the present application, there is provided a construction method of an extended image sequence, including: acquiring tomographic data of a target projection object; Performing tomographic image reconstruction processing on the tomographic data to generate a tomographic reconstructed image sequence of the target projection object, wherein the tomographic reconstructed image sequence of the target projection object is used for representing a maximum image sequence reconstructed under the current reconstruction condition; And constructing an extended image sequence of the simulated projection object matched with the target projection object based on the tomographic reconstructed image sequence of the target projection object, wherein the extended image sequence is used for representing the extended image sequence of the simulated projection object which is not reconstructed under the current reconstruction condition and needs to be constructed virtually. Preferably, before the acquiring of the tomographic data of the target projection object, the method further includes: obtaining flat image data of the target projection object; The constructing an extended image sequence of a simulated projection object matched with the target projection object based on the tomographic reconstructed image sequence of the target projection object specifically comprises the following steps: and constructing an extended image sequence of a simulated projection object matched with the target projection object based on the tomographic reconstruction image sequence of the target projection object and the flat image data of the target projection object. Preferably, the constructing an extended image sequence of a simulated projection object matched with the target projection object based on the tomographic reconstructed image sequence of the target projection object and the flat image data of the target projection object specifically includes: establishing a compression mapping model based on a tomographic reconstruction image sequence of the target projection object and flat image data of the target projection object; And carrying out reconstruction processing based on the compressed mapping model and the flat image data of the target projection object to obtain an expanded image sequence. Preferably, the method further comprises: and correcting the extended image sequence based on the tomographic reconstructed image sequence of the target projection object to obtain a corrected extended image sequence, wherein the correction comprises size and/or definition adjustment. . Preferably, the constructing an extended image sequence of a simulated projection object matched with the target projection object based on the tomographic reconstructed image sequence of the target projection object and the flat image data of the target projection object includes: Acquiring an equivalent phantom of the target projection object, wherein the equivalent phantom comprises positions corresponding to the tomographic scanning sequence and the expansion sequence; The tomographic reconstruction image sequence of the target projection object and the flat image data of the target pr