In a recent study published in Radiology by Xu et al, a novel artificial intelligence (AI) model was found to accurately predict the risk of death from lung cancer, cardiovascular disease, and other causes. The researchers used data from low-dose computed tomography (CT) scans of the lungs for their predictions. Currently, the U. S Preventive Services Task Force recommends annual lung screenings with low-dose CT scans for individuals aged 50 to 80 who are at high risk of lung cancer, such as heavy smokers. These scans not only provide information about the lungs but also about other structures in the chest, including body composition. Previous research has shown that abnormal body composition, such as obesity and loss of muscle mass, is linked to chronic health conditions like metabolic disorders. Body composition has also been found to be useful for risk stratification and prognosis in cardiovascular disease and chronic obstructive pulmonary disease. Furthermore, it has been discovered that body composition can impact survival and quality of life in lung cancer therapy. Lead study author Kaiwen Xu explains that while the primary focus of CT scans is to identify lung cancer nodules, there is additional valuable anatomical information, such as body composition, coded in the images. Xu and his colleagues have developed an AI algorithm that can automatically derive body composition measurements from low-dose CT scans of the lungs.

In their recent study, the researchers analyzed over 200, 000 CT scans from participants in the National Lung Screening Trial to evaluate the potential benefits of AI-derived body composition measurements. The inclusion of these measurements in the patients' assessments led to improved risk predictions for lung cancer mortality, cardiovascular disease mortality, and all-cause mortality. Particularly, measurements related to myosteatosis, which is now considered a more predictive indicator of health outcomes than reduced muscle bulk, were found to be strong predictors of mortality. Xu suggests that CT scans, originally ordered for lung cancer detection purposes, contain a wealth of additional information, including body composition and coronary artery calcification, which relates directly to cardiovascular disease risk. Consequently, using body composition measurements from low-dose CT scans of the lungs could offer opportunities for opportunistic screenings and potentially aid physicians in routine clinical settings. Xu emphasizes that automatic AI body composition analysis extends the value of lung screenings beyond early lung cancer detection, allowing for the identification of high-risk patients who may benefit from interventions like physical conditioning or lifestyle modifications, even in the early stages before disease onset. The researchers hope to further explore the benefits of their AI model in future studies with extended follow-up periods that investigate how changes in body composition correspond to health outcomes. Please note that the content of this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO) and does not necessarily reflect their ideas and opinions.