Skip to content

Diagnosis and post-surgical prognosis of pulmonary sub-solid nodules using deep learning: medical imaging evidence inside the model

Diagnosis and post-surgical prognosis of pulmonary sub-solid nodules using deep learning: medical imaging evidence inside the model

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2000029908
Enrollment
Unknown
Registered
2020-02-16
Start date
2020-03-01
Completion date
Unknown
Last updated
2020-02-17

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

Conditions

Lung cancer

Interventions

Gold Standard:Pathological examination and immunohistochemical staining.
Index test:Classification&#32
prediction&#32
of&#32
deep&#32
neural&#32
network.

Sponsors

Shanghai General Hospital
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. To include 1000 cases of subsolid pulmonary nodules with pathological classification: there is no limit to the time. We have done chest CT examination in two hospital areas and found sub solid pulmonary nodules (< 3cm). Within one month, we have carried out surgery, frozen pathology in the excised tissues, paraffin section and immunohistochemical staining to obtain pathological classification results. Gender is not limited. 2. To include 500 cases of pulmonary nodules with postoperative follow-up results: before March 1, 2017 (three years ago), chest CT examination was performed in two hospital areas of our hospital to find sub solid pulmonary nodules (< 3cm). Within one month, surgery was performed, and frozen pathology was performed on the excised tissues, and paraffin section and immunohistochemical staining were performed to obtain pathological classification results. After March 1, 2020, follow-up CT examination will be carried out to find out whether there is recurrence. Gender is not limited.

Exclusion criteria

Exclusion criteria: 1. Aged < 18 years. 2. The image of pulmonary nodule is not clear, so it is difficult to segment with surrounding tissue.

Design outcomes

Primary

MeasureTime frame
Diagnostic accuracy;F1 score;

Countries

China

Contacts

Public ContactXueqian Xie

Shanghai General Hospital

adrianxie@163.com+86 13564412266

Outcome results

None listed

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