Multi-organ sketching method and device
A multi-organ and organ technology, applied in the field of medical devices, can solve the problem of low radiotherapy treatment efficiency, and achieve the effect of improving radiotherapy treatment efficiency, high accuracy, and fast organ delineation.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0052] figure 1 It is a schematic flowchart of a multi-organ delineation method provided in Embodiment 1 of the present invention, as shown in figure 1 As shown, the method may include the following steps:
[0053] Step 1: Construct a multi-category training set containing historical data of multiple patients, wherein the historical data of each patient includes historical 3D CT image data of the patient and corresponding multi-organ delineation data.
[0054] Step 2: using a multi-classification training set containing historical data of multiple patients to train a convolutional neural network for synchronously delineating multiple organs to obtain a trained convolutional neural network for synchronously delineating multiple organs;
[0055] Step 3: Using the trained convolutional neural network for synchronously delineating multiple organs, perform multi-organ synchronous delineation processing on the 3D CT image data to be delineated, and obtain a 3D CT image containing d...
Embodiment 2
[0064] figure 2 It is a schematic structural diagram of a multi-organ delineation device provided in Embodiment 2 of the present invention, as shown in figure 2 As shown, the device may include a multi-organ synchronous delineation training set building module, a multi-organ synchronous delineation model training module and a multi-organ synchronous delineation module, wherein:
[0065] 1. Multi-organ synchronous delineation training set building block
[0066] The multi-organ synchronous delineation training set construction module is used to construct a multi-classification training set containing historical data of multiple patients, wherein the historical data of each patient includes the patient's historical 3D CT image data and corresponding multi-organ delineation data .
[0067] Specifically, the multi-organ synchronous delineation training set construction module can determine multiple organs that need to be delineated simultaneously according to actual needs; obt...
Embodiment 3
[0078]In actual work, doctors need to outline multiple organs, and most of the existing technologies are single-organ models, that is, running a model can only predict the corresponding single organ area, so if you need to outline multiple organs, you need to run multiple organs For the corresponding single-organ model, running the single-organ model corresponding to multiple organs at the same time will consume a lot of video memory, and running the single-organ model in batches will take longer.
[0079] Convolutional neural network (CNN) in deep learning is a common method for interpreting images, and has a large number of example applications in the fields of image classification, semantic segmentation, object detection, and image understanding. U-shaped CNN (or UNeT) extracts abstract features through multi-layer downsampling of CT images, and then upsamples back to the original size to realize the delineation of organ regions. The convolutional neural network for synchro...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


