Skip to content

Multi-type Artifact Removal and Electron Density Reconstruction for Radiotherapy Images Based on Unpaired CT-CBCT and Dual-Encoder Physics-Informed Neural Networks (Dual-PINN)

Multi-type Artifact Removal and Electron Density Reconstruction for Radiotherapy Images Based on Unpaired CT-CBCT and Dual-Encoder Physics-Informed Neural Networks (Dual-PINN)

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600128842
Enrollment
Unknown
Registered
2026-07-27
Start date
2026-07-30
Completion date
Unknown
Last updated
2026-08-03

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Various malignant tumors requiring radiotherapy

Interventions

Retrospective CBCT Imaging Cohort:None
Unpaired Planning CT Reference Cohort:None

Sponsors

Peking University Shenzhen Hospital
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: 1.CBCT data collected during the treatment of patients with tumors in various body regions; 2. Planning CT data (unpaired) for patients with tumors in various regions; 3. Each case must include at least 50 consecutive slices; 4. The DICOM format is complete, containing pixel data and all necessary metadata.

Exclusion criteria

Exclusion criteria: 1. The file is damaged or the pixel data cannot be read. 2. Only a single slice or the number of slices is less than 50; 3. There are severe motion artifacts in the image, which completely prevent the identification of the anatomical structures.

Design outcomes

Primary

MeasureTime frame
Consistency of radiotherapy dose distribution (Gamma pass rate, relative error of target D95);

Secondary

MeasureTime frame
Synthetic CT image quality (PSNR, SSIM, MAE, RMSE);

Countries

China

Contacts

Public ContactShuo Yu

Peking University Shenzhen Hospital

ys0519@outlook.com+86 755 8392333

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Aug 10, 2026