The Chest X- Ray: The Ship has Sailed, But Has It?

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Underwriting Science

Background:

• Chest X-rays (CXRs) were historically used in life insurance underwriting to assess mortality risk.
• CXRs were abandoned due to applicant inconvenience, cost, and limited predictive value.
• Advancements in deep learning present an opportunity to re-evaluate the utility of CXRs.

Purpose of the Study:

• To investigate the potential of deep learning techniques to enhance the value of chest X-rays (CXRs) in modern life insurance underwriting.
• To determine if CXR data, analyzed with AI, can provide additional insights into mortality risk assessment.

Main Methods:

• Review of historical underwriting practices involving CXRs.
• Exploration of current deep learning methodologies applicable to medical image analysis.
• Assessment of the feasibility of integrating AI-powered CXR analysis into underwriting workflows.

Main Results:

• Deep learning models show promise in extracting novel predictive features from CXR images.
• AI-driven analysis could potentially overcome limitations of traditional CXR interpretation.
• The reintroduction of CXRs, enhanced by AI, may offer improved mortality risk stratification.

Conclusions:

• Deep learning offers a pathway to revive the utility of chest X-rays in life insurance underwriting.
• AI-powered CXR analysis has the potential to significantly improve the accuracy of mortality risk assessment.
• Further research is warranted to validate the predictive power of AI-analyzed CXRs in underwriting.