Neural Network Low-Dose CT Image Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computed tomography (CT) images reconstructed from low-dose data often suffer from noise and artifacts, such as staircase artifacts, which degrade image quality and can impact medical diagnosis, while high-dose CT scans improve quality but at the cost of increased radiation exposure.
Innovation Solution
A method using neural network models, specifically Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), or Generative Adversarial Networks (GAN), to convert low-dose CT images into high-dose images by training models on paired low-dose and high-dose data, reducing noise and enhancing image quality without increasing radiation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a high-dose CT scan is performed, then image quality is improved, but radiation exposure increases
Solution Approach 1:
The patent creates a virtual copy of high-dose image quality by training a neural network model on paired low-dose and high-dose images. The model learns to map low-dose images to their high-dose equivalents, generating synthetic high-dose images without actual high-dose radiation exposure. This copying approach preserves the diagnostic quality of high-dose images while eliminating the harmful radiation exposure.
2Object-affected harmful factors
If a low-dose CT scan is performed, then radiation exposure is reduced, but image quality deteriorates due to noise and artifacts
Solution Approach 1:
The patent introduces a neural network model as an intermediary between low-dose images and high-dose images. This intermediary processes the noisy low-dose images and transforms them into high-quality images that resemble high-dose images. The model acts as a mediator that compensates for the quality loss in low-dose images by learning the transformation patterns from training data, effectively removing noise and artifacts while preserving diagnostic information.
3Productivity
If conventional reconstruction algorithms are used on low-dose data, then processing speed is maintained, but image quality suffers from noise and staircase artifacts
Solution Approach 1:
The patent replaces conventional mechanical reconstruction algorithms with an intelligent neural network-based system. Instead of using traditional iterative or analytical reconstruction methods that produce noisy images with staircase artifacts, the system employs a pre-trained neural network that has learned optimal transformation patterns. This substitution maintains fast processing speeds while dramatically improving image quality by replacing the mechanical reconstruction process with an intelligent mapping approach.
Data Source
AI summary
A method may include obtaining a first set of projection data with respect to a first dose level; reconstructing, based on the first set of projection data, a first image; determining a second set of projection data based on the first set of projection data, the second set of projection data relating to a second dose level that is lower than the first dose level; reconstructing a second image based on the second set of projection data; and training a first neural network model based on the first image and the second image. In some embodiments, the trained first neural network model may be configured to convert a third image to a fourth image, the fourth image exhibiting a lower noise level and corresponding to a higher dose level than the third image.