Mathematics
Scientific paper
May 2002
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2002stin...0244432t&link_type=abstract
Technical Report, NASA/TP-2002-210781; S-891; NAS 1.60:210781
Mathematics
Aeroembolism, Decompression Sickness, Goodness Of Fit, Random Processes, Parameter Identification, Mathematical Models, Censored Data (Mathematics), Denitrogenation, Extravehicular Activity, Data Correlation
Scientific paper
Venous gas emboli (VGE) (gas bubbles in venous blood) are associated with an increased risk of decompression sickness (DCS) in hypobaric environments. A high grade of VGE can be a precursor to serious DCS. In this paper, we model time to Grade IV VGE considering a subset of individuals assumed to be immune from experiencing VGE. Our data contain monitoring test results from subjects undergoing up to 13 denitrogenation test procedures prior to exposure to a hypobaric environment. The onset time of Grade IV VGE is recorded as contained within certain time intervals. We fit a parametric (lognormal) mixture survival model to the interval-and right-censored data to account for the possibility of a subset of "cured" individuals who are immune to the event. Our model contains random subject effects to account for correlations between repeated measurements on a single individual. Model assessments and cross-validation indicate that this limited failure population mixture model is an improvement over a model that does not account for the potential of a fraction of cured individuals. We also evaluated some alternative mixture models. Predictions from the best fitted mixture model indicate that the actual process is reasonably approximated by a limited failure population model.
Chhikara Raj S.
Conkin Johnny
Powell Michael R.
Thompson Laura A.
No associations
LandOfFree
Modeling Grade IV Gas Emboli using a Limited Failure Population Model with Random Effects does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Modeling Grade IV Gas Emboli using a Limited Failure Population Model with Random Effects, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Modeling Grade IV Gas Emboli using a Limited Failure Population Model with Random Effects will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-834074