The integration of AI and low-dose CT scans has the potential to revolutionize risk prediction for diseases like lung cancer and cardiovascular disease. By expanding the utility of these scans to include body composition measurements, AI can help identify high-risk individuals for early intervention. This early detection and intervention approach can significantly improve patient outcomes and ultimately save lives.
Artificial intelligence (AI) has the potential to revolutionize risk prediction for various diseases, including lung cancer and cardiovascular disease. A recent study published in the Radiology journal by the Radiological Society of North America (RSNA) has shown that AI can utilize data from low-dose CT scans of the lungs to enhance risk prediction for death from these diseases and other causes. These findings have significant implications for early detection and intervention strategies.
Currently, the U.S. Preventive Services Task Force recommends annual lung screening with low-dose CT scans for individuals aged 50 to 80 years who are at high risk of developing lung cancer, such as longtime smokers. These CT scans not only capture images of the lungs but also provide valuable information about other structures in the chest.
“When we’re looking at the CT images, the primary focus is on identifying nodules suspicious for lung cancer, but there is much more anatomical information coded in the space, including information on body composition.”
This statement by Kaiwen Xu, the lead author of the study and a Ph.D. candidate in the Department of Computer Science at Vanderbilt University, emphasizes the untapped potential of these CT scans. Xu and his colleagues have developed an AI algorithm that can derive body composition measurements from lung screening LDCT scans. Body composition refers to the percentage of fat, muscle, and bone in the body. Abnormal body composition, such as obesity or loss of muscle mass, is closely linked to chronic health conditions like metabolic disorders.
Moreover, studies have demonstrated that body composition is a useful tool for risk stratification and prognosis in cardiovascular disease and chronic obstructive pulmonary disease. In the context of lung cancer therapy, body composition has also been shown to impact survival and quality of life. Therefore, it is imperative to explore the potential of AI-derived body composition measurements in improving risk prediction for various diseases.
Xu’s study assessed the added value of AI-derived body composition measurements by analyzing the CT scans of over 20,000 individuals from the National Lung Screening Trial. The results were promising, as the inclusion of these measurements significantly enhanced risk prediction for death from lung cancer, cardiovascular disease, and overall mortality.
According to Xu, “Automatic AI body composition potentially extends the value of lung screening with low-dose CT beyond the early detection of lung cancer. It can help us identify high-risk individuals for interventions like physical conditioning or lifestyle modifications, even at a very early stage before the onset of disease.”
Furthermore, measurements related to fat within muscle were found to be particularly strong predictors of mortality. This aligns with existing research, which suggests that the infiltration of skeletal muscle with fat, known as myosteatosis, may be more predictive of health outcomes than reduced muscle bulk.
The use of opportunistic screening, where imaging for one purpose provides information about other conditions, holds immense potential in routine clinical practice. In the case of lung screening LDCT, the practice of deriving body composition measurements from these scans is a prime example. The images captured during lung cancer screening contain abundant information beyond just the detection of lung cancer. They can also provide insights into body composition or coronary artery calcification, which is directly associated with cardiovascular disease risk.
This study, however, focused solely on baseline screenings. In future research, Xu and his team aim to investigate the changes in body composition over time and how they correlate with health outcomes. A longitudinal study that follows individuals over a period of time would offer valuable insights into the progression and impact of body composition changes.
The integration of AI and low-dose CT scans has the potential to revolutionize risk prediction for diseases like lung cancer and cardiovascular disease. By expanding the utility of these scans to include body composition measurements, AI can help identify high-risk individuals for early intervention. This early detection and intervention approach can significantly improve patient outcomes and ultimately save lives.
Source: Radiological Society of North America
Journal reference: Xu, K., et al. (2023) AI Body Composition in Lung Cancer Screening: Added Value Beyond Lung Cancer Detection. Radiology. doi.org/10.1148/radiol.222937.
