Abstract :

In order to enhance outcomes, it is important to discover lung cancer early. It is critical to enhance the ability of chest CT scans to identify nodules. Prior AI modules have shown promising results, leading to better performance in recognizing lung lesions in some datasets. Having said that, their false-positive (FP) rate is really high. No definitive evidence of its efficacy in clinical practice exists at this time. The goal of our AI-assisted CT scans was to reduce FP. Procedures and materials: Obtaining CT scans from sixty patients was the goal. This research aimed to identify lung nodules in five senior physicians who were not told about these patients. Without the use of AI, two medical professionals identified and labeled lung nodules by hand. Three other physicians identified and classified lung nodules with the help of AI before interpreting the results by hand. The artificial intelligence software uses a deep learning architecture. The findings show that 266 nodules were det