A Prospective, Non-interventional Cohort Study of Subsolid Pulmonary Nodules

Prediction of Growth Trend of Pulmonary Nodules Based on Multi-temporal and Multimodal Medical Data and Its Impact on Pulmonary Nodule Management Strategies

Some studies have shown that the model for judging and predicting the growth of sub-solid pulmonary nodules through big data and deep learning can detect nodule growth earlier. Since most of the training data come from large foreign samples, most of the validated data are CT data from a single center or a few centers, and their generalization ability needs to be further verified. In order to better study subsolid pulmonary nodules in the lungs in China, we plan to conduct a prospective, multicenter, non-interventional observational cohort study.

Through the follow-up of pulmonary nodules, artificial intelligence based on CT was used to study the natural evolution process of subsolid pulmonary nodules, as well as the development law and prognosis of pulmonary subsolid nodules under treatment or no treatment according to clinical guidelines.

Inclusion Criteria:

• Age≥ 18 years old, gender is not limited
• Confirmed subsolid pulmonary nodule by CT, initial pulmonary nodule ≥ 5 mm (including subsolid components) and ≤ 30 mm, do not shrink or disappear after 3 months Follow-up.
• ECOG 0～2
• Able to provide CT in DICOM format
• Informed consent obtained

Exclusion Criteria:

• Long-term use of immunosuppressants and hormone drugs
• History of previous malignancy (if there is no recurrence or carcinoma in situ for more than 5 years after receiving curative therapy, it can be enrolled)
• Inability to cooperate with regular follow-up
• Poor general condition, with the life expectancy less than 6 months