5 Key Takeaways

• 1

  An AI model demonstrated comparable accuracy to radiologists in estimating pleural effusion volume on chest radiography.

• 2

  Moderate agreement was observed between all three approaches and computed tomography, with substantial estimation errors.

• 3

  Mean estimation errors were 300 mL for radiologist 1, 250 mL for radiologist 2, and 264 mL for AI.

• 4

  Radiologist 2 outperformed AI in identifying pleural effusions, while performance did not differ statistically between radiologist 1 and AI.

• 5

  The study's limitations included its retrospective nature, single-institution setting, and potential spectrum bias affecting generalizability.