3. Vai trò của ứng dụng trí tuệ nhân tạo vindr trong dự báo nguy cơ ác tính của nốt phổi đơn độc

Abstract

Solitary pulmonary nodule was defined as well-circumscribed round lesion measuring up to 3 cm in diameter and surrounded by aerated lung without pulmonary collapse, mediastinal lymph nodes or pleural effusion. This is a common pulmonary lesion, which can have many causes. Benign causes include: tuberculoma, harmatoma, pulmonary lymph nodes, sarcoidosis, aspergilloma. Malignant etiologies include lung cancer, metastatic lung cancer, lymphoma, etc. Artificial intelligence in diagnosing isolated lung nodules aims at early detection of lesions, accurate diagnosis and prognosis of disease. Our study was conducted to evaluate the role of VinDr artificial intelligence application in predicting the individual risk of solitary pulmonary nodules in order to  assist clinicians  for acquiring additional means  for timely diagnosis and  provide appropriate treatment options for the patient. This is a  descriptive study  composed of 23 patients with isolated pulmonary nodules undergoing diagnostic transthoracic lung biopsy ; AI analysis results showed  a sensitivity of 83.3%, a specificity  of 100%, and a negative predictive value  of 62.5%. Moreover, there was a good consensus between AI analysis and histopathological results. In conclusion, our study suggests that AI has the potential to become an essential tool in predicting the malignant risk of Solitary pulmonary nodule - SPN. However, more research is needed to validate the use of AI in clinical practice, and a combination of AI and traditional methods is recommended for more accurate diagnosis and management of SPN.