Skip to content

Research on Intelligent Diagnosis of X-ray Advantaged Diseases and Structured Report Generation Based on Deep Learning and Radiomics

Research on Intelligent Diagnosis of X-ray Advantaged Diseases and Structured Report Generation Based on Deep Learning and Radiomics

Status
Recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2400092216
Enrollment
Unknown
Registered
2024-11-12
Start date
2024-05-01
Completion date
Unknown
Last updated
2024-11-18

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

Conditions

Pneumothorax, pleural effusion, pneumothorax with hemothorax, rib fracture

Interventions

Gold Standard:Reference standards established by senior radiologists
Index test:Evaluating common diseases detected by X-ray using Convolutional Neural Networks (CNNs) and generating structured reports using Natural Language Processing (NLP) techniques

Sponsors

Shanghai Changzheng Hospital
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. Inclusion criteria for healthy volunteers: There are no pneumothorax, pleural effusion, acute abdomen and other X-ray dominant system diseases and common diseases. 2. Inclusion criteria for people with pneumothorax, pleural effusion, acute abdomen and related diseases: (1) The quality of the plain film image is qualified: no artifacts, good sharpness, good contrast and clarity; (2) There are clear X-ray plain film or CT image diagnostic results.

Exclusion criteria

Exclusion criteria: (1) Poor quality of plain films: blurry, underexposed or overexposed, severe artifacts; (2) Plain radiographs and other imaging examinations have no diagnostic results; (3) Those who are pregnant or lactating; (4) Uncontrolled epilepsy or psychiatric illness that cannot cooperate with X-ray photography; (5) Patients with cough, sputum production or difficulty breathing and cannot inhale and hold breath normally.

Design outcomes

Primary

MeasureTime frame
Receiver Operating Characteristic Curve Area Under the Curve;

Countries

China

Contacts

Public ContactFan Li

Shanghai Changzheng Hospital

fanli0930@163.com+86 135 6468 4699

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 4, 2026