Artificial Intelligence: Can It Save Lives, Hospitals, and Lung Screening?

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Early lung cancer detection is crucial for patient survival.
• Incidental pulmonary nodules on imaging present a management challenge due to high volume.
• Artificial intelligence (AI) offers a potential solution for detecting and triaging these nodules.

Purpose of the Study:

• To evaluate an AI tool's ability to read imaging reports and identify indeterminate pulmonary nodules.
• To assess if AI can triage nodules without additional personnel.
• To determine if AI can aid in early lung cancer diagnosis and improve survival rates.

Main Methods:

• An incidental lung nodule clinic (ILNC) was established using AI software and a nurse practitioner.
• The AI system analyzed radiology reports, identifying lung nodules greater than 3 mm.
• Indeterminate nodules were referred to the ILNC for further evaluation and management.

Main Results:

• AI analyzed 502,632 imaging reports, identifying 22,136 (4.4%) with positive findings.
• 518 patients with indeterminate nodules were referred to the ILNC.
• 14 lung cancers were detected (2.9%), with 8 cases at Stage I, leading to timely treatment.

Conclusions:

• AI software can effectively supplement healthcare practitioners in managing incidental pulmonary nodules.
• The AI tool facilitates earlier lung cancer diagnosis, potentially saving lives.
• Implementing AI in radiology reporting can generate value-based revenue and improve patient care.