Stanford Bone Bayes is a Bayesian network that models clinical and radiographic inputs to compute diagnosis, differential diagnosis, and probabilities.

Read the Stanford Bone Bayes publication here.

Here is a tutorial on how to select features.

Christopher Beaulieu MD PhD ( link )
Bao Do MD ( link )


Age ( years ) :

Gender : male female

# of lesions : solitary multiple

Bone location :

Longitudinal location :

Proximal or distal :

Transverse location :

Central vs eccentric : central eccentric not applicable

Density :

Matrix / texture :

Transition zone / border :

Cortex :

Periosteum :

Lesion to shaft ratio : 0-25% 25-50% 50-75% 75-100% > 100% not applicable

Physis : closed open

Expansion : non-expansile expansile

Associated soft tissue mass : yes no

Associated pathologic fracture : yes no


Knowledge source : Stanford Jones, Dahlin, Campanacci, Pai & Yap literature review (July 2020)



What is your top (1-3) differential diagnosis ? ( Please help us train the A.I. )




Stanford Bone Bayes (c) 2016-2020, Stanford Department of Radiology