Clinical trial · Observational
Developing a Machine Learning Model to Predict Pleural Adhesion Preoperatively Using Pleural Ultrasound
Developing a Machine Learning Model to Predict Pleural Adhesion Preoperatively Using Pleural Ultrasound: A Prospective Observational Study
- Source
- ClinicalTrials.gov
- Retrieved
- Sep 8, 2026
- Layer
- normalized (units and labels harmonized; values unchanged)
- Run
- ING-CLINICALTRIALS-20260908-000001
Summary
Brief summary (as posted)
This study aims to investigate the accuracy of using pleural ultrasound (USP) to identify pleural adhesions in patients who plan to receive video-assisted thoracoscopic surgery. It employs three-dimensional convolutional neural network (3D-CNN) technology to process USP-related images and video data for machine learning, and to establish a diagnostic model for identifying pleural adhesions using 3D-CNN-USP. The study will determine the sensitivity, specificity, positive predictive value, and negative predictive value of 3D-CNN-USP in identifying pleural adhesions. Additionally, it will explore the feasibility and effectiveness of using 3D-CNN-USP for preoperative identification of pleural adhesions in VATS, thereby supporting the implementation of day surgery in thoracic surgery and ultimately serving clinical practice.
Conditions
Conditions (3)
Free-text conditions as registered, with the CancerIndex entity they were reconciled to and the match type.
| Condition (as posted) | Mapped entity | Match | Confidence |
|---|---|---|---|
| Lung Cancer | Malignant Lung Neoplasm | CURATED_EXACT | 0.92 |
| Machine Learning | — | UNRESOLVED | — |
| Pleural Diseases | — | UNRESOLVED | — |
Interventions
Interventions (1)
| Intervention | Type | Mapped drug | Match |
|---|---|---|---|
| Pleural ultrasound | Diagnostic Test | — | UNRESOLVED |
Design
Arms and outcomes
Arms (1)
- label
- Pleural ultrasound group
- description
- Patients who accept pleural ultrasound preoperatively.
- interventionNames
- Diagnostic Test: Pleural ultrasound
Primary outcomes (1)
- measure
- Sensitivity
- timeFrame
- From enrollment to the end of surgery.
- description
- Sensitivity of three-dimensional convolutional neural network (3D-CNN) in identifying pleural adhesions using pleural ultrasound (USP). The sensitivity value ranges from 0 to 100, with higher values indicating greater sensitivity.
Secondary outcomes (3)
- measure
- Specificity
- timeFrame
- From enrollment to the end of surgery.
- description
Eligibility
Eligibility (as posted)
- Sex
- All
- Minimum age
- 12 Years
- Maximum age
- 80 Years
Show eligibility criteria text
Inclusion Criteria: 1\. Patients who plan to accept VATS surgery. Exclusion Criteria: 1. Patients who can not obtain detailed clinical information; 2. Patients or their family members who can not understand the conditions and objectives of the study or refuse to participate in the study; 3. Patients with conditions affecting observation, such as skin lesions, infections, or scars in the area of the chest wall to be examined.
References
Publications (2)
- RESULTMason AC, Miller BH, Krasna MJ, White CS. Accuracy of CT for the detection of pleural adhesions: correlation with video-assisted thoracoscopic surgery. Chest. 1999 Feb;115(2):423-7. doi: 10.1378/chest.115.2.423. PMID 10027442
- RESULTCassanelli N, Caroli G, Dolci G, Dell'Amore A, Luciano G, Bini A, Stella F. Accuracy of transthoracic ultrasound for the detection of pleural adhesions. Eur J Cardiothorac Surg. 2012 Nov;42(5):813-8; discussion 818. doi: 10.1093/ejcts/ezs144. Epub 2012 Apr 19. PMID 22518039