Sybil: A Validated Deep Learning Model to Predict Future Lung Cancer Risk From a Single Low-Dose Chest Computed Tomography

Abstract

PURPOSE Low-dose computed tomography (LDCT) for lung cancer screening is effective, although most eligible people are not being screened. Tools that provide personalized future cancer risk assessment could focus approaches toward those most likely to benefit. We hypothesized that a deep learning model assessing the entire volumetric LDCT data could be built to predict individual risk without requiring additional demographic or clinical data. METHODS We developed a model called Sybil using LDCTs from the National Lung Screening Trial (NLST). Sybil requires only one LDCT and does not require clinical data or radiologist annotations; it can run in real time in the background on a radiology reading station. Sybil was validated on three independent data sets: a heldout set of 6,282 LDCTs from NLST participants, 8,821 LDCTs from Massachusetts General Hospital (MGH), and 12,280 LDCTs from Chang Gung Memorial Hospital (CGMH, which included people with a range of smoking history including nonsmokers). RESULTS Sybil achieved area under the receiver-operator curves for lung cancer prediction at 1 year of 0.92 (95% CI, 0.88 to 0.95) on NLST, 0.86 (95% CI, 0.82 to 0.90) on MGH, and 0.94 (95% CI, 0.91 to 1.00) on CGMH external validation sets. Concordance indices over 6 years were 0.75 (95% CI, 0.72 to 0.78), 0.81 (95% CI, 0.77 to 0.85), and 0.80 (95% CI, 0.75 to 0.86) for NLST, MGH, and CGMH, respectively. CONCLUSION Sybil can accurately predict an individual's future lung cancer risk from a single LDCT scan to further enable personalized screening. Future study is required to understand Sybil's clinical applications. Our model and annotations are publicly available.

Nanoprobe-based molecular imaging for tumor stratification

“…377,378 For instance, the combining of low-dose computed tomography (LDCT) and a new algorithm can help to stratify tumor patients and select small cell subpopulations from heterogeneous data for specific treatments. 379…”

Tissue-engineered models of lung cancer premalignancy

“…The clearest form of lung cancer prevention is to reduce smoking prevalence, but more than half of all lung cancer cases are diagnosed in former smokers. In fact, a new deep learning model can now accurately predict lung cancer risk in former smokers from a single computed tomography (CT) scan 3 . This innovation underscores the need for interventions targeting premalignant lesions in these high-risk individuals.…”

The Intersection of Lung Cancer Screening, Radiomics, and Artificial Intelligence: Can One Scan Really Predict the Future Development of Lung Cancer?

“…In the article that accompanies this editorial, Mikhael et al 9 report that an artificial intelligence and deep learning model, called Sybil, may predict an individual's future lung cancer risk after one baseline computed tomography chest scan. This model is an important first step toward a precision approach to lung cancer screening, but understanding who would truly benefit from this technology will require significantly more investment in prospective studies targeting groups with differing risk profiles.…”

“…8 Although clinics, hospitals, and health systems struggle with how to improve screening uptake, adherence, and the tracking of suspicious findings, researchers have been focusing on using clinical prediction modeling, harnessing the power or the electronic medical record, radiomics, and using artificial intelligence (AI) to aid in improving outcomes all along the screening continuum. This brings us to the study that accompanies this editorial by Mikhael et al, 9 which explores the use of a validated deep learning model to predict future lung cancer risk from a single LDCT. Before delving into the paper itself, we should take a step back and briefly review terms such as AI and deep machine learning (DL) which are increasingly referred to in the medical literature but are often misused and misunderstood by the medical community.…”