EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Integrating Deep Learning of Low-Dose CT Imaging With Clinical Data for Lung Cancer Risk Prediction

Renzo Phellan Aro, Stephen Lam, Matthew T Warkentin, Geoffrey Liu, Brenda Diergaarde, David O Wilson, Jian-Min Yuan, Hamad Al-Sawaihey, Kiera R Murison, Elham Khodayari-Moez, Yonathan Brhane, Rafael Meza und 2 weitere

CHEST · 2026

DOI: 10.1016/j.chest.2026.02.022 ↗

Kurz gefasst

Fachartikel in der Zeitschrift CHEST, 2026. Verfasst von Renzo Phellan Aro und 13 weiteren Autor:innen. Beteiligte Einrichtung: Prosserman Centre for Population Health Research, Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, ON, Canada. Hauptthemen laut MEDLINE: Lung Neoplasms, Deep Learning, Tomography, X-Ray Computed, Early Detection of Cancer. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: NCI NIH HHS, Canadian Cancer Society, National Institutes of Health National Cancer Institute.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND: Low-dose CT (LDCT) imaging 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. RESEARCH QUESTION: Can the integration of DL based on LDCT scans and clinical data improve lung cancer risk prediction? STUDY DESIGN AND METHODS: Retrospective cohort data from 4 different screening programs, 1 used for model training and 3 used for external validation. Data were collected between 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 of smoking 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 epidemiologic factors were evaluated for their added predictive value. RESULTS: This analysis used 52,482 LDCT scan 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, 0.81-0.85) compared with Sybil (AUC, 0.80; 95% CI, 0.78-0.82) in year 6. The difference is most notable when nodules are absent. Sybil-Epi's AUC was 0.76 (95% CI, 0.70-0.82) and Sybil's AUC was 0.64 (95% CI, 0.57-0.70). INTERPRETATION: Our results show that 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 DL and clinical and epidemiologic factors significantly improved model predictive performance.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Lung NeoplasmsDeep LearningTomography, X-Ray ComputedEarly Detection of CancerHumansFemaleRetrospective StudiesMaleMiddle AgedRisk AssessmentRadiation DosageAgedPredictive Value of TestsRisk FactorsROC Curve

Stichwörter der Autor:innen

deep learninglow-dose CTlung cancerrisk prediction

Autor:innen und Institutionen

  • Prosserman Centre for Population Health Research, Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, ON, Canada.
  • 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,…
  • 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; Cancer Epidemiology and Prevention Program, University of Pittsburgh Medical Center, Pittsburgh, PA.
  • Cancer Epidemiology and Prevention Program, University of Pittsburgh Medical Center, Pittsburgh, PA; Department of Medicine, University of Pittsburgh, School of Medicine, Pittsburgh, PA.

Förderung

  • NCI NIH HHS
  • Canadian Cancer Society
  • National Institutes of Health National Cancer Institute

Zitieren

APA

Phellan Aro, R., Lam, S., Warkentin, M. T., Liu, G., Diergaarde, B., Wilson, D. O., Yuan, J. M., Al-Sawaihey, H., Murison, K. R., Khodayari-Moez, E., Brhane, Y., Meza, R., Myers, R., & Hung, R. J. (2026). Integrating Deep Learning of Low-Dose CT Imaging With Clinical Data for Lung Cancer Risk Prediction. CHEST. https://doi.org/10.1016/j.chest.2026.02.022

Vancouver

Phellan Aro R, Lam S, Warkentin MT, Liu G, Diergaarde B, Wilson DO, et al. Integrating Deep Learning of Low-Dose CT Imaging With Clinical Data for Lung Cancer Risk Prediction. CHEST. 2026. doi:10.1016/j.chest.2026.02.022

BibTeX
@article{PhellanAro2026,
  author = {Phellan Aro, Renzo and Lam, Stephen and Warkentin, Matthew T and Liu, Geoffrey and Diergaarde, Brenda and Wilson, David O and Yuan, Jian-Min and Al-Sawaihey, Hamad and Murison, Kiera R and Khodayari-Moez, Elham and Brhane, Yonathan and Meza, Rafael and Myers, Renelle and Hung, Rayjean J},
  title = {Integrating Deep Learning of Low-Dose CT Imaging With Clinical Data for Lung Cancer Risk Prediction},
  journal = {CHEST},
  year = {2026},
  doi = {10.1016/j.chest.2026.02.022}
}
Kostenlos starten →