Lung cancer ct ai studies focus mostly on detection tasks
Disentangling Lung-Cancer CT/LDCT AI: A Systematic Evidence Map of Clinical Tasks, Evidence Chains, and Translational Gaps
Machine Learning
Summary
Lung cancer diagnosis using CT scans is an important medical task, and artificial intelligence (AI) has been applied to help with this. The authors found that most AI studies focus on detecting cancer rather than predicting its future risk, which is a different medical challenge. They also discovered only a small number of studies provide complete and reliable evidence for clinical use, highlighting many gaps before AI tools can be fully trusted for patients. The study maps and categorizes these AI efforts to clarify what has been done and where more work is needed.
What this means in practice
- •For hospital data teams: Organize and prioritize AI studies for clinical lung cancer detection to focus validation efforts on reliable and actionable tools.
- •For medical device developers: Develop AI systems for lung cancer CT scans with an emphasis on improving future risk prediction capabilities beyond simple detection.
A survey. It maps existing work.
Authors
Surajit Das
Abstract
Artificial-intelligence studies using computed tomography (CT) for lung cancer are often broadly labelled "prediction" despite addressing clinically distinct tasks. We systematically mapped CT/low-dose CT (LDCT)-centered lung-cancer AI using five-database retrieval, full-text eligibility assessment, role-aware modality/omics extraction, clinical-task classification, and a Multi-Tier Evidence Graph (MTEG). The final corpus comprised 293 studies (2016-2026): 230 Detection, 8 future Risk-prediction, and 55 Other studies. Clinical variables (96.2%), 3D CT/LDCT (73.0%), and radiomics (63.5%) predominated, whereas external validation (29.0%), calibration (20.5%), decision-curve analysis (13.0%), longitudinal CT (17.7%), and saliency/attribution XAI (21.5%) were less frequent. The MTEG comprised 377 nodes and 3,444 edges; only 31 studies (10.6%) completed the six-tier substantive evidence chain, with greatest attrition at reasoning/explanation. Overall, the literature is detection-dominated, genuine future risk prediction remains uncommon, and complete translational evidence chains are rare.