Integrating deep learning of low-dose computed tomography with clinical data for lung cancer risk prediction.
Authors
Affiliations (13)
Affiliations (13)
- Prosserman Centre for Population Health Research. Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, ON, Canada. Electronic address: [email protected].
- Department of Respiratory Medicine, University of British Columbia, Vancouver, BC, Canada; Department of Integrative Oncology, British Columbia Cancer Research Institute, Vancouver, BC, Canada.
- Prosserman Centre for Population Health Research. Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, ON, Canada; Department of Oncology, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada.
- Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada; Department of Medical Oncology and Hematology, Princess Margaret Cancer Centre, Toronto, ON, Canada.
- Department of Human Genetics, University of Pittsburgh School of Public Health, Pittsburgh, PA, USA; Cancer Epidemiology and Prevention Program, UPMC Hillman Cancer Center, Pittsburgh, PA, USA.
- Cancer Epidemiology and Prevention Program, UPMC Hillman Cancer Center, Pittsburgh, PA, USA; Department of Medicine, University of Pittsburgh, Pittsburgh, PA, USA.
- Cancer Epidemiology and Prevention Program, UPMC Hillman Cancer Center, Pittsburgh, PA, USA; Department of Epidemiology, University of Pittsburgh School of Public Health, Pittsburgh, PA, USA.
- Royal Cornwall Hospitals NHS Trust, Truro, UK.
- Prosserman Centre for Population Health Research. Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, ON, Canada; Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada.
- Prosserman Centre for Population Health Research. Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, ON, Canada.
- Department of Integrative Oncology, British Columbia Cancer Research Institute, Vancouver, BC, Canada.
- Department of Respiratory Medicine, University of British Columbia, Vancouver, BC, Canada.
- Prosserman Centre for Population Health Research. Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, ON, Canada; Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada. Electronic address: [email protected].
Abstract
Low-dose computed tomography (LDCT) screening reduces lung cancer mortality, the leading cause of cancer deaths globally. Segmentation-free deep learning (DL) models such as Sybil can improve screening efficiency but require extensive validation and possible improvement. Can the integration of deep learning based on LDCT scans and clinical data improve lung cancer risk prediction? Retrospective cohort data from 4 different screening programs, one used for model training and three for external validation. Data collected between the years 2002 and 2021. The median follow-up period was 7 years. All participants had a history of either current or former smoking, with at least 10 pack-years or who smoked over 20 years. The area under the receiver operating characteristic curve (AUC) was calculated for lung cancer risk within 1 to 6 years, stratified by pulmonary nodule presence and size. Key clinical and epidemiological factors were evaluated for their added predictive value. This analysis uses 52,482 LDCT series from 22,469 participants. Sybil's AUC ranged from 0.93 in year 1 and reduced to 0.79 in year 6 in the independent cohorts. The predictive performance was suboptimal in the absence of documented nodules (AUC=0.64), and for small nodules (AUC=0.61) in year 6. Our new model, Sybil-Epi, trained with baseline scans, achieved higher predictive performance (AUC=0.83, 95% CI 0.81 to 0.85) compared to Sybil (AUC=0.80, 95% CI 0.78 to 0.82) in year 6. The difference is most notable when nodules are absent, with Sybil-Epi AUC of 0.76 (95% CI 0.70 to 0.82) and Sybil AUC of 0.64 (95% CI 0.57 to 0.70). Sybil performs better for short-term lung cancer risk, but the predictive accuracy was suboptimal when nodules were absent. Our integrated Sybil-Epi model with deep learning and clinical-epidemiological factors significantly improved model predictive performance.