Application of artificial intelligence analysis software to assess pulmonary small nodules in patients with osteosarcoma

Abstract Objective To explore the clinical application value of artificial intelligence analysis software in the assessment of pulmonary small nodules in patients with osteosarcoma. Methods From March 2011 to February 2022, 138 patients with osteosarcoma diagnosed in the Department of Oncology, Hangzhou Third People's Hospital underwent chest thin-section CT and pulmonary nodule screening based on artificial intelligence analysis software.The composition, quantity, distribution and dynamic changes of pulmonary nodules were analyzed to determine whether they were metastases. Results 849 CT scans were performed in 138 patients with osteosarcoma. Artificial intelligence software detected 3989 pulmonary nodules, of which 3069 (76.9%) were small nodules with a diameter of ≤ 5 mm, of which 1749 (57.0%) were solid nodules, 650 (21.2%) were calcified nodules. 127 cases (92%,127/138) had small nodules, 23 cases (18.1%, 23/127) had dynamic changes in small nodules, 16 cases (69.5%, 16/23) were diagnosed with lung metastases, There was no significant difference in the location of pulmonary distribution between the diameter ≤ 5 mm and the diameter > 5 mm nodules. Conclusion Artificial intelligence diagnostic technology helps to intelligently evaluate the size, number and density of small pulmonary nodules in patients with osteosarcoma, and dynamic monitoring helps to characterize small pulmonary nodules.