Determining the likelihood of malignancy in solitary pulmonary nodules with Bayesian analysis. Part II. Application

Radiology. 1993 Feb;186(2):415-22. doi: 10.1148/radiology.186.2.8421744.


Four board-certified radiologists estimated the probability of malignancy in 66 cases of solitary pulmonary nodules. Two other radiologists evaluated the same nodules according to various radiographic and clinical findings. These findings were then used to estimate the probability of malignancy by using previously derived likelihood ratios and the Bayes theorem. The readers using Bayesian analysis performed significantly better than the expert readers (P < .05) when individual radiographs were considered and when all radiologic studies were combined. In addition, the readers using Bayesian analysis misclassified fewer malignant nodules as benign (mean, 6.5) than did the expert readers (mean, 6.5) than did the expert readers (mean, 16.5). The authors conclude that Bayesian analysis may be a useful aid in the evaluation of solitary pulmonary nodules.

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Bayes Theorem*
  • False Negative Reactions
  • False Positive Reactions
  • Female
  • Humans
  • Male
  • Middle Aged
  • ROC Curve
  • Radiography
  • Risk Factors
  • Smoking
  • Solitary Pulmonary Nodule / diagnostic imaging
  • Solitary Pulmonary Nodule / epidemiology*
  • Solitary Pulmonary Nodule / pathology