Low-Dose CT Lung Nodule Detection via Deep Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current health examination methods, such as full-dose computed tomography (CT) scans, expose patients to significant radiation, making it desirable to develop methods for processing low-dose CT images to detect lung nodules and coronary artery calcification (CAC) with reduced radiation exposure while maintaining diagnostic accuracy.
Innovation Solution
The method involves processing low-dose CT images using deep learning models like U-Net and Efficient Net for lung nodule detection and CAC scoring, respectively, to classify lung nodules based on radiomics features and generate reports with treatment recommendations, thereby reducing radiation exposure while maintaining diagnostic accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full-dose CT scan is used for health examination, then diagnostic accuracy is improved, but radiation exposure increases
Solution Approach 1:
The patent changes the radiation dose parameter from full-dose to low-dose CT scanning, and uses deep learning models to compensate for the reduced image quality, thereby maintaining diagnostic accuracy while reducing radiation exposure
Solution Approach 2:
The patent introduces deep learning models (U-Net, Efficient Net) as intermediary tools that process low-dose CT images to enhance diagnostic accuracy, acting as a mediator between low-dose imaging and accurate diagnosis
2Object-affected harmful factors
If low-dose CT method is used, then radiation exposure is reduced, but image quality deteriorates
Solution Approach 1:
Deep learning models serve as intermediary processing tools that take low-dose CT images as input and output enhanced images or direct diagnostic results, compensating for the quality loss from reduced radiation dose
Solution Approach 2:
The patent changes the radiation dose parameter and uses computational algorithms to compensate, transforming the physical imaging parameter change into a computational problem that can be solved through deep learning
3Reliability
If multiple examinations are performed to ensure diagnostic accuracy, then diagnostic reliability is improved, but examination time increases
Solution Approach 1:
The patent makes the low-dose CT examination perform multiple diagnostic functions simultaneously (lung nodule detection, CAC scoring, etc.) through deep learning analysis, replacing the need for multiple separate examinations
Solution Approach 2:
The patent combines multiple diagnostic tasks (lung nodule detection, classification, CAC scoring) into a single integrated low-dose CT examination workflow, reducing the number of separate examinations needed
Data Source
AI summary
Disclosed are methods and devices of processing a low-dose computed tomography (CT) image. The present disclosure provides a method of processing a low-dose CT image. The method comprises: receiving a first chest image; receiving a first chest image; detecting at least one lung nodule in the first chest image; determining at least one lung nodule region of the first chest image based on the at least one lung nodule; and classifying the at least one lung nodule region based on a first set of radiomics features of the at least one lung nodule region of the first chest image to obtain a nodule score of the at least one lung nodule in the lung nodule region. The first chest image generated by a low-dose CT method.