Pure ground glass nodules have a low overall malignancy risk of <10% and can also be risk assessed using predictive modeling based on individual and nodule specifics. Lung nodules can be categorized according to density into solid nodules, pure ground glass nodules, and mixed nodules with both solid and ground glass components. Of these, mixed nodules have the highest risk of malignancy, and pure ground-glass nodules have the lowest risk of malignancy, which is generally below 10%. Nodules in pure ground glass nodules vary in size, morphology, and margins, and patients have different malignant risks depending on factors such as age, gender, and family history. The classical prediction models commonly used are Herder, Brock, VA, and Mayo models. In addition, prediction can also be based on biological markers such as carcinoembryonic antigen and genomics. For example, the Brock model predicts risk based on the patient's gender, age, family history of lung cancer, presence or absence of emphysema, as well as the location, type, size, number, and burr sign of the nodule. Generally those with low malignant risk (<10%) can be followed up, while those with high malignant risk require treatment such as puncture biopsy or surgical resection. Patients with pure ground-glass nodules are advised to consult a specialist for risk assessment based on the individual as well as the lesion, and to follow medical advice.