A New Method to Differentiate Benign and Malignant Pulmonary Nodules by Mass Spectrometry Combined With Artificial Intelligence.
试验速览
- 阶段
- 不适用
- 状态
- 招募中
- 发起方
- 入组人数
- 150
- 试验地点
- 1
- 主要终点
- Establish an model for differentiating lung nodules.
研究概览
简要总结
The goal of this observational clinical trial is to establish a new method for differentiating benign and malignant pulmonary nodules by peripheral blood detection in patients with pulmonary nodules (<3cm). The main questions it aims to answer is: How to combine blood metabolomic mass spectrometry detection and artificial intelligence image analysis to establish a new model for differentiating benign and malignant pulmonary nodules. Participants will be asked provide 4 mL peripheral blood for the test.
详细描述
The aim of this clinical trial is to establish a new method for differentiating benign and malignant pulmonary nodules by the combination of metabolomics analysis and artificial intelligence (AI) analysis. It is expected to improve the accuracy of the identification of benign and malignant pulmonary nodules. Patients with clinical suspected malignant pulmonary nodules will be included in this trial. The subjects will be divided into three group by CT image presentation: (1) pure ground-glass nodule (pGGN), (2) part-solid nodule (PSN), (3) solid nodule (SN). Peripheral blood of subjects will be collected and detected by mass spectrometry to obtain the metabolomic characterization. The classification model of each group will be constructed based on the data analysis algorithm by machine learning. The diagnostic efficacy of the new model combined with the AI image analysis system for differentiating benign and malignant pulmonary nodules will be analyzed.
研究设计
- 研究类型
- Observational
- 观察模型
- Cohort
- 时间视角
- Prospective
入排标准
- 年龄范围
- 18 Years 至 —（Adult, Older Adult）
- 性别
- All
- 接受健康志愿者
- 否
入选标准
- •Age ≥18 years;
- •CT imaging shows the presence of pulmonary nodule <3cm which is scheduled for puncture biopsy or surgery (i.e., the target lesion), the presence of ≥2 target lesions of the same type (categorized by density) are allowed;
- •Subjects are in good condition with Eastern Cooperative Oncology Group (ECOG) scale of 0-2;
- •Subjects with fair vital organ function, defined as: white blood cells ≥3.0×10^9/L, platelets ≥75×10^9/L, hemoglobin ≥90g/L, alanine aminotransferase and aspartate aminotransferase ≤2.5 times the upper limit of normal values, and serum creatinine <178μmol/L;
- •Subjects must have the ability to understand and sign the informed consent in writing voluntarily.
排除标准
- •Imaging examination have suggested the possibility of metastasis at other sites;
- •≥2 target lesions with different type categorized by density;
- •History of malignant disease;
- •Severe vascular lesions within the last 3 months, or known significant active infection, during acute/chronic tuberculosis infection, or severe cardiovascular and cerebrovascular diseases, dysfunction of liver and renal, or significant endocrine and metabolic disorders, or other serious concomitant diseases that are not controlled;
- •The specialist/surgeon assessed that puncture or surgery is not available, with contraindication such as coagulation disorders, cardiorespiratory insufficiency, etc.;
- •History of uncontrolled epilepsy, central nervous system disease, or psychiatric disorders, that may affect the signing of informed consent;
- •Pregnant or breastfeeding women;
- •Other conditions deemed by the investigator to be unsuitable for enrollment.
结局指标
主要结局
Establish an model for differentiating lung nodules.
时间窗: 6 months
To establish a new method for differentiating benign and malignant pulmonary nodules by the combination of metabolomics analysis and artificial intelligence (AI) image analysis.
次要结局
未报告次要终点
研究者
Guangying Zhu
chief physician
China-Japan Friendship Hospital