Usage of Yb(OH)CO3 Nanoparticles-based Computed Tomography Image in the Prediction Model of Lung Biopsy Pneumothorax

Yb(OH)CO3 Nanoparticles-based Computed Tomography Image

Abstract

This study aimed to explore the application of Yb(OH)CO3 nanoparticles in the prediction model of computed tomography (CT) contrast lung biopsy pneumothorax (LBP) by synthesizing a new type of Yb(OH)CO3 nanoparticles, which provides a basis for clinical application. Yb(OH)CO3 nanoparticles were prepared by a homogeneous precipitation process based on urea, and the samples were characterized by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Methyl thiazolyl tetrazolium (MTT) assay was adopted in vitro cytotoxicity test, optical density (OD) value was measured with a microplate reader at a wavelength of 570 nm, the cell viability was analyzed, and the cytomorphological changes were observed with an optical microscope. Besides, 40 selected patients with CT-guided percutaneous needling biopsy (PTNB) were rolled into a control group and an observation group. A pneumothorax prediction model was constructed under the support vector machine (SVM) model. The accuracy, sensitivity, and specificity of SVM for pneumothorax prediction were 88.7%, 71.3%, and 100%, respectively; while those of CT imaging with Yb(OH)CO3 nanoparticles were 96.7%, 84.3%, and 100%, respectively. Mean squared errors (MESs) of Yb(OH)CO3 nanoparticles-based CT contrast agent and conventional contrast agent were 14,532 and 7,021, respectively, and mutual information (MI) was 0.1232 and 0.2354, respectively, so the difference between the two groups was significant (P < 0.05). In summary, Yb(OH)CO3 nanoparticles-based CT contrast agents showed good histocompatibility and low toxicity, and the Yb(OH)CO3 nanoparticles-based CT images showed high accuracy in the prediction of LBP.