Computers can undertake complex calculations and iterative actions much faster than humans so the mathematical calculation of the probabilities of diagnoses and outcomes (and their utilities) can be undertaken. Since we do not normally undertake these calculations, they add an extra dimension to clinical evaluation
and can be used to shed light on the significance of features of the history, clinical examination and investigation.
These methods have not been widely used in orthopaedics
. Decision analysis has been reported in the diagnosis of disc herniation
, mortality in trauma
, resurfacing the patella in TKA, outcome of the contralateral side in Slipped Capital Femoral Epiphysis
, acute Achilles Tendon rupture
, open and arthroscopic Bankart repair
, DDH
, osteoporosis
, and OA of the wrist
. There is more extensive literature on the related technique of cost effectiveness, which has been recently reviewed by Brauer et al.
Lanny Johnson has made use of his enormous but humbling experience in knee diagnosis to produce the website FreeKneeDx. In the backgound information provided on the website he notes that clinical diagnosis was no better than 80% accurate even when backed by extensive experience. He also found that knees described as 'normal' by their owners had significant symptomatology when questioned in detail
. The website administers a history questionaire on knee symptoms and calculates the probability of any of about 40 diagnoses using Discriminant Function analysis
and offers 85% accuracy according to this study.
h3 References
