Combined model integrating pericardial fat and tumor radiomics with clinical data to classify pulmonary nodules: A multicentric analysis
Principal Investigator
Name
Wenjuan Huang
Degrees
M.D.
Institution
Harbin medical university cancer hospital
Position Title
no
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-1346
Initial CDAS Request Approval
Nov 5, 2024
Title
Combined model integrating pericardial fat and tumor radiomics with clinical data to classify pulmonary nodules: A multicentric analysis
Summary
Preoperative discrimination of benign and malignant pulmonary nodules is critical for informing clinical management decisions. Recent studies revealed that pericardial fat plays a key role in cancer initiation and progression. This study aimed to develop a combined model that integrates pericardial fat and tumor radiomics features with clinical data to classify pulmonary nodules as benign or malignant. Patients with pulmonary nodules were enrolled from multiple medical centers. Radiomics features were extracted from both three-dimensional (3D) pericardial fat and tumor on preoperative computed tomography (CT) images. These features, along with clinical predictors, were integrated into a hybrid model. The model’s performance was validated in test and validation cohorts based on accuracy, discrimination, clinical benefits, and generalization ability. Performance improvements in incorporating fat features for the model were assessed using the net reclassification index (NRI) and integrated discrimination improvement (IDI).
Aims
Develop a combined model that integrates pericardial fat and tumor radiomics features with clinical data to classify pulmonary nodules as benign or malignant.
Collaborators
Rui-tao wang: Department of Internal Medicine, Harbin Medical University Cancer Hospital, Harbin Medical University.