Thyroid nodule calculator

Age (required):  
Sex (required):  
Microcalcification:  
Taller-than-wide:  
Margin:    
Capsular invasion:  
Architecture:          
 
Echogenicity:    
Ring-down artifact:  
Vascularity:    

Probability of papillary thyroid cancer:

When you select a single feature, the calculator gives you the probability from Bayes theorem. When you have entered in all of the features, it gives you adjusted naive Bayes. With an intermediate number of features, it gives you both naive Bayes and adjusted naive Bayes (presumably the actual probability is between these two values).

Important limitations
This is a research and educational tool, and should not be used to make medical decisions. The model only applies to papillary thyroid cancer. Many important factors, such as family history, history of neck irradiation, size, growth, and presence of a multinodular goiter are not included in this model. This model has not been validated on children. The probabilities may be different for different patient populations.

References
Liu YI, Kamaya A, Desser TS, and Rubin DL. (2011) "A Bayesian Network for Differentiating Benign From Malignant Thyroid Nodules Using Sonographic and Demographic Features." AJR 196(5): W598-W605.

Boas FE. (2011) "Calculating the probability of malignancy of thyroid nodules using adjusted naive Bayes."