adenocarcinoma
Conditions
Interventions
Gold Standard:Paraffin pathological results
Index test:Model for predicting pathological high-risk factors of lung cancer based on graph convolutional neural network
Sponsors
Beijing Chao-Yang Hospital, Capital Medical University
Eligibility
Sex/Gender
All
Age
18 Years to 80 Years
Inclusion criteria
Inclusion criteria: Inclusion criteria for the first registration:(1) Chest computed tomography (CT) scan fulfilled all of the following conditions: (a) high-resolution thin-layer CT or enhanced CT; (b) CT scan layer thickness less than 1.5 mm; (c) imaging consideration for lung adenocarcinoma (LUAD). (2) The tumor characteristics also met the following conditions: (a) maximum tumor diameter = 3 cm; (b) no pleural invasion was seen. (3) Patient's age was >18 years. (4) ECOG classification of 0 or 1. (5) Signed written informed consent. Those who met the above criteria were enrolled in the first part of the training group prospective study. Inclusion criteria for the second registration: preoperative criteria: (1) each organ function could tolerate the surgery; (2) preoperative body temperature =38°C. Intraoperative criteria: (1) intraoperative frozen return of histologically confirmed invasive lung adenocarcinoma; (2) no pleural invasion; (3) no intrathoracic metastasis.
Exclusion criteria
Exclusion criteria: Exclusion criteria: Patients were excluded preoperatively if they met any of the following criteria: (1) Preoperative imaging consideration of lymph node metastases; (2) Active bacterial or fungal infection; (3) Prior history of other malignancies (except for clinically cured malignancies, e.g., papillary thyroid cancer, breast cancer, renal clear cell carcinoma, etc.) (4) Women who are pregnant or breastfeeding; (5) Patients undergoing preoperative adjuvant radiotherapy or neoadjuvant therapy; (6) Psychiatric disorders;
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Accuracy of a predictive model for pathological high-risk factors of lung cancer based on graph convolutional neural networks; | — |
Secondary
| Measure | Time frame |
|---|---|
| Comparison of accuracy of lung cancer pathological high-risk factors prediction model based on graph Convolutional neural network and artificial interpretation results of intraoperative frozen pathology;Sensitivity, specificity, accuracy; | — |
Countries
China
Contacts
Public ContactJiYing
Beijing Chao-Yang Hospital, Capital Medical University
Outcome results
None listed