A new deep-learning AI algorithm significantly lowered false positives in lung nodule malignancy assessment while maintaining high detection rates.
Key Details
- 1Deep-learning algorithm estimated lung nodule malignancy risk using LDCT data.
- 2Trained on 16,077 nodules (1,249 malignant) from the NLST; validated on a pooled cohort of 4,146 participants and 7,794 nodules.
- 3AUC for AI model was equal or superior to PanCan: ~0.98 (1yr), 0.96 (2yr) in pooled cohort; up to 0.95 for indeterminate nodules.
- 4For size-matched nodules, AI model AUC was 0.79 vs. PanCan at 0.6.
- 5At 100% sensitivity for cancer, AI flagged 68.1% of benign nodules as low-risk vs. PanCan's 47.4% (39.4% relative reduction in false positives).
- 6Authors stress the need for prospective clinical validation before routine use.
Why It Matters
Reducing false positives in lung cancer screening can minimize unnecessary procedures, patient anxiety, and healthcare costs. Improved AI-based risk stratification stands to improve the efficiency and effectiveness of LDCT lung cancer screening programs.
Source
AuntMinnie
Related News
•HealthExec
New Report Highlights Clinical AI Performance, Sustainability, and Adoption Challenges
A multi-institutional review details key challenges, progress, and sustainability concerns in deploying clinical AI in real-world healthcare settings.
•Cardiovascular Business
FDA Clears AI Platform for Comprehensive Cardiac Risk Assessment on CT
HeartLung Corporation's AI-CVD receives FDA clearance for opportunistic multi-condition screening on routine chest CT scans.
•AuntMinnie
LLM Boosts Terminology Expansion in Radiology Reports Over RadLex
A large language model (LLM) significantly outperforms RadLex in expanding terms for radiology report language standardization.