We have developed a distance education system for developing skills in endoscopic paranasal sinus surgery, to enable efficient remote training of novices in manual skills such as their standing position and posture, and the insertion angle/depth and holding of surgical instruments. The system uses a precise model of human paranasal sinuses and the HyperMirror (HM) telecommunication interface. HM is a virtual mirror allowing clear visualization of differences in manual operation between the trainee and remote expert. This paper outlines the proposed system and describes remote training experiments between two locations 200 miles apart. In the experiments, two expert surgeons trained 17 novices for 40 to 60 min on probing of the nasofrontal duct and aspiration of the maxillary sinus, and subjectively evaluated their manual skills. The results showed that most of the novices improved their manual skills and were able to complete each procedure.

For Bayesian inference on the mixture of factor analyzers, natural conjugate priors on the parameters are introduced, and then a Gibbs sampler that generates parameter samples following the posterior is constructed. In addition, a deterministic estimation algorithm is derived by taking modes instead of samples from the conditional posteriors used in the Gibbs sampler. This is regarded as a maximum a posteriori estimation algorithm with hyperparameter search. The behaviors of the Gibbs sampler and the deterministic algorithm are compared on a simulation experiment.