Potential added value of an AI software with prediction of malignancy for the management of incidental lung nodules

Abstract

Purpose: To determine the impact of an artificial intelligence software predicting malignancy in the management of incidentally discovered lung nodules. Materials and methods: In this retrospective study, all lung nodules ≥ 6 mm and ≤ 30 mm incidentally discovered on emergency CT scans performed between June 1, 2017 and December 31, 2017 were assessed. Artificial intelligence software using deep learning algorithms was applied to determine their likelihood of malignancy: most likely benign (AI score 75%). Predictions were compared to two-year follow-up and Brock's model. Results: Ninety incidental pulmonary nodules in 83 patients were retrospectively included. 36 nodules were benign, 13 were malignant and 41 remained indeterminate at 2 years follow-up.AI analysis was possible for 81/90 nodules. The 34 benign nodules had an AI score between 0.02% and 96.73% (mean = 48.05 ± 37.32), while the 11 malignant nodules had an AI score between 82.89% and 100% (mean = 93.9 ± 2.3). The diagnostic performance of the AI software for positive diagnosis of malignant nodules using a 75% malignancy threshold was: sensitivity = 100% [95% CI 72%-100%]; specificity = 55.8% [38–73]; PPV = 42.3% [23–63]; NPV = 100% [82–100]. With its apparent high NPV, the addition of an AI score to the initial CT could have avoided a guidelines-recommended follow-up in 50% of the benign pulmonary nodules (6/12 nodules). Conclusion: Artificial intelligence software using deep learning algorithms presents a strong NPV (100%, with a 95% CI 82–100), suggesting potential use for reducing the need for follow-up of nodules categorized as benign.

Keywords

• Lung nodule
• Lung cancer
• Artificial intelligence
• Radiomics
• Deep learning