Skip to content

Construction and clinical translation of a predictive model for high-risk pathological subtypes and gene mutations in lung adenocarcinoma based on the fusion of artificial intelligence and radiomics features

Construction and clinical translation of a predictive model for high-risk pathological subtypes and gene mutations in lung adenocarcinoma based on the fusion of artificial intelligence and radiomics features

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600128628
Enrollment
Unknown
Registered
2026-07-23
Start date
2026-08-01
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

Lung adenocarcinoma

Interventions

Modeling team and validation team:None

Sponsors

The Second Affiliated Hospital of Hainan Medical University
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: 1. Age: Age at surgery >=18 years old; 2. Diagnosis: Patients received surgical resection, and primary lung adenocarcinoma was confirmed by postoperative paraffin pathology; 3. Imaging data: Contrast-enhanced chest CT was performed in our hospital before surgery (generally within 1 week), and the original thin-slice (slice thickness <= 3 mm) DICOM-format imaging data are complete and available; 4. Molecular pathological data: Test results of EGFR gene mutation status based on surgically resected specimens (test methods may include amplification refractory mutation system-PCR, next-generation sequencing, etc.); 5. Pathological subtype data: Complete postoperative pathological reports containing information on predominant pathological subtypes classified according to the 2015 WHO classification criteria or the IASLC/ATS/ERS multidisciplinary classification (especially the identification of high-risk subtypes such as micropapillary and solid subtypes); 6. Baseline data: Accessible basic clinical information including at least gender, age and smoking history.

Exclusion criteria

Exclusion criteria: 1. Image quality issues: Preoperative CT images have artifacts that severely affect lesion observation and delineation (such as respiratory motion artifacts and metal implant artifacts), excessively low image resolution, or the scan range does not fully cover the entire lung. 2. Previous treatment history: Any form of anti-tumor therapy administered for the pulmonary lesion before surgery, including but not limited to puncture biopsy, radiofrequency ablation, radiotherapy, chemotherapy, targeted therapy or immunotherapy. 3. Interference from comorbidities: Previous or concurrent history of malignant tumors in other organs (excluding basal cell carcinoma of the skin). 4. Incomplete information: Missing key data, such as absence of postoperative pathological reports, unavailable EGFR test results, lack of preoperative enhanced CT images, or severely incomplete clinical baseline data precluding analysis.

Design outcomes

Primary

MeasureTime frame
EGFR gene mutation status;

Secondary

MeasureTime frame
Pathological subtypes of lung adenocarcinoma;

Countries

China

Contacts

Public ContactPeng Dongge

The Second Affiliated Hospital of Hainan Medical University

1453779508@qq.com+86 898 7881931

Outcome results

None listed

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