Introduction: Verification of new decompression procedures has traditionally been based on observing the occurrence of decompression sickness (DCS) in test dives. Several hundred exposures are required to determine the safety of a procedure with any degree of certainty. The number of venous gas emboli (VGE) corresponds with the risk of getting DCS and detection of VGE has been used as an alternative method for validation of decompression procedures. We propose a new and improved method for validation based on detection of VGE.
Methods: Our Bayesian statistics method combines results from ultrasound detection of VCE in test dives with knowledge about the correspondence between VGE and DCS risk obtained from a large number of previous experimental studies. Our algorithm is implemented in a computer program; it estimates DCS risk and 95% credible intervals for the tested procedure.
Results: We have applied the method to available VGE data from tested air diving procedures with between 7 and 14 test dives for each procedure. The estimated credible intervals correspond to confidence intervals from 130-250 dives using the binomial distribution of the traditional "DCS observation validation."
Discussion: We conclude that, compared with previous methods, the proposed method can greatly reduce the number of dives required to validate or reject new decompression procedures.