Radiology, Vascular Diseases
Conditions
Brief summary
In this study, patients with chest pain, lung cancer, pulmonary embolism, and routine inpatient physical examination were selected as the research objects, and the experimental design of retrospective cohort study was adopted to carry out artificial intelligence analysis related to pulmonary vascular diseases in patients with multi-dimensional big data. The multi-modal CT acquisition process included plain scan CT(NCCT) and CT pulmonary angiography (CTPA). Ctpa-like image effects can be simulated or reconstructed by non-enhanced plain scan CT images, so that CTPA-like image quality can be obtained without injecting contrast agent. The synthetic CTPA images were further analyzed by artificial intelligence to assist doctors in the intelligent diagnosis of pulmonary vascular diseases.
Detailed description
A non-enhanced plain scan CT image simulates or reconstructs an image effect similar to that of CTPA through the following technical solutions: 1. Data acquisition: Obtain plain scan CT image data of the examined person, including multiple layers of image slices. 2. Image preprocessing: Preprocessing of plain scan CT images, including denoising, enhancing contrast and other steps, to improve image quality and lay the foundation for subsequent processing. 3. Vascular segmentation: Advanced image segmentation algorithms, such as the deep learning-based segmentation method, are used to segment the vascular structure from the preprocessed plain scan CT images. The key to this step is to accurately identify and extract vascular areas while reducing interference from non-vascular tissue. 4. Blood vessel enhancement: For the segmented blood vessel structure, a specific image enhancement algorithm is used to enhance blood vessels to make them clearer and more continuous. 5. Image synthesis: The enhanced vascular image is fused with the original plain scan CT image to generate the final CTPA image. During the synthesis process, the contrast between blood vessels and surrounding tissues can be adjusted as needed to optimize the display effect. 6. Post-processing and evaluation: Post-processing of synthesized CTPA images, such as smoothing, artifact removal, etc., and quality assessment to ensure that the images meet the needs of clinical diagnosis.
Interventions
DIAGNOSTIC_TESTDeep learning imaging enhancement
Conventional imaging or down-sampling imaging from CT or MR are enhanced by approved deep learning method.
Sponsors
Xin Lou
Study design
Observational model
CASE_ONLY
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
18 Years to 100 Years
Healthy volunteers
No
Inclusion criteria
* Age ≥18≤100 years old Scan the pulmonary artery and its major branches Patients with suspected pulmonary embolism who received CTPA had a set of CTPA and CT scans The image quality meets the requirements of diagnosis and post-processing Patients who completed the examination in accordance with the data collection criteria Clinical data and follow-up were complete
Exclusion criteria
* Age <18 years or age >100 years The image is incomplete or incorrect Pulmonary artery absent or underenhanced Severe motion artifacts or image noise affect evaluation of pulmonary embolism History of aortic reconstruction, replacement, or stent implantation Congenital variations in the whole or important branches of the aorta in adults (e.g. bovine aortic arch, abnormal right subclavian artery) Severe hypovolemia and hemodynamic instability Severe heart failure with low ejection fraction Dialysis patient
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| The performance of deep enhanced imaging in lesion detection and diagnosis | 2 year | The performance of deep enhanced imaging in lesion detection and diagnosis, including imaging quality, accuracy, sensitivity and specificity in lesion detection and imaging diagnosis. |
Contacts
Primary ContactLing Liu
Outcome results
None listed