Deco Theory 101, 201, 301, and 401

Please register or login

Welcome to ScubaBoard, the world's largest scuba diving community. Registration is not required to read the forums, but we encourage you to join. Joining has its benefits and enables you to participate in the discussions.

Benefits of registering include

  • Ability to post and comment on topics and discussions.
  • A Free photo gallery to share your dive photos with the world.
  • You can make this box go away

Joining is quick and easy. Log in or Register now!

Then that's likely the difference.

thank you, makes sense. From the Parker et.al. (1996) paper I still think that ignoring O2 in the ISS integral would be more appropriate and won't change the ranking order of the four profiles, or even result in larger differences. Do you know if in the ISS calculation for these graphs, were separate tissues used for oxygen with own half-times, or did they just add oxygen to nitrogen tissues?


Hello,

Two apologies.

First, for being so late in responding to this. I have had a very busy few weeks clinically.

Second, for mislabelling the Y axis on the ISS graphs. They have been transposed between presentations so many times I had not noticed that the units were not properly defined any more.

Stuart, to your question.

You describe a somewhat hypothetical question that could not happen in reality - particularly your first dive profile. It would not be possible to have one tissue at 10 and all the others at 1 because in achieving 10 in one tissue would inevitably mean that other tissues with similar kinetics would be much closer, with a graduated change to differing values as the tissue half times diverged. Moreover comparing profiles with such divergent supersaturation patterns like the hypothetical you propose is not the intent of ISS. This is a bit similar to the argument Ross tried when suggesting that the ISS was crap because going on a ski trip at altitude would produce a total ISS similar to a dive. ISS comparisons are only useful for comparing profiles that are substantially equivalent in important respects (eg to the same depth for the same bottom time with the same total decompression time, but with different distribution of stop depths and times). It would not be a valid method of comparing two profiles that would produce the sort of wildly different patterns of supersaturation that you describe in your hypothetical.

How you actually use ISS, and the extent to which we can consider it a validated means of evaluating optimal decompression remain open to discussion, but it has been an interesting metric in comparing the profiles discussed in debates over deep vs less deep stop approaches.

Simon M

Thank you very much for the explanation. In

Survanshi, S. & Weathersby, Paul & Thalmann, E.. (1996). Statistically Based Decompression Tables X: Real-Time Decompression Algorithm Using a Probabilistic Model. NMRI 96-06

they use a model of DCS risk from ISS values in an interesting way: by a *weighted* sum of the ISS of the compartments. So it considers that slow and fast compartments have different sensitivities to ISS. The resulting risk value however should be comparable between widely different dives (different depth, different bottom times, different total run times). Many of this group's papers use this approach, also for the linear-exponential kinetics. Do you know if this probabilistic approach based on ISS is still popular, or did they abandon this approach?
 
thank you, makes sense. From the Parker et.al. (1996) paper I still think that ignoring O2 in the ISS integral would be more appropriate and won't change the ranking order of the four profiles, or even result in larger differences. Do you know if in the ISS calculation for these graphs, were separate tissues used for oxygen with own half-times, or did they just add oxygen to nitrogen tissues?
It sounds like you're ignoring partial pressures from O2, CO2, and H2O (water vapor). I don't think that's the best model. VPM itself uses the other gases in its model (e.g. look in the VPM code for "Constant_Pressure_Other_Gases"). Thalmann has also stated, "The total tissue gas burden is the sum of the inert gas burden and the tissue O2, CO2, and water vapor partial pressures."

These gases are modeled as constant additions to compartment pressures, not as independent gases with half-times, etc.
 
It sounds like you're ignoring partial pressures from O2, CO2, and H2O (water vapor). I don't think that's the best model. VPM itself uses the other gases in its model (e.g. look in the VPM code for "Constant_Pressure_Other_Gases"). Thalmann has also stated, "The total tissue gas burden is the sum of the inert gas burden and the tissue O2, CO2, and water vapor partial pressures."

These gases are modeled as constant additions to compartment pressures, not as independent gases with half-times, etc.


Thanks, you're right that the Thalmann LE (linear-exponential) adds a constant to the compartment pressures. Doing so makes the numbers closer to those given in the presentation.

I then tried to minimize ISS and found: the shallower the better. GF150/50 is very good in terms of ISS but will almost certainly cause a hit. I think the problem with ISS is the following: We're summing up the 16 ISS values of the 16 compartments without any weighting; whereas the probabilistic Thalmann models carefully fit weights of the compartments. Without weights, the choice of discrete compartment half-times matters a lot. The Bühlmann compartments are narrowly spaced on the fast side (5,8,12.5,...) and widely spaced on the slow side (390,498,635). That's good for the Bühlmann model looking for the maximum supersaturation only; but when you sum up supersaturation, spreading half-times in a different way would yield totally different ISS results, and may even turn the tables in favour of the deep stops if you'd include more fast compartments and fewer slow ones.
As long as we don't have weights for the compartments, looking at the maximum of the 16 compartment ISS values instead of their sum may make more sense. Minimizing max(ISS) still results in shallow profiles but not as extreme as minimizing sum(ISS).

About treating O2 as an independent gas: that's from a different Thalmann paper where they find that their ISS prediction model underestimates P(DCS) for dives using 100% O2 as a decompression gas. They tried various models and the only one that improved the prediction was a model with a separate compartment for O2 with a very short half-time of 0.4min:
Parker, E. & Survanshi, S. & Thalmann, E. & Weathersby, Paul. (1996). Statistically Based Decompression Tables IX: Probabilistic Models of the Role of Oxygen in Human Decompression Sickness. NMRI 96-05, March 1996.
 
I then tried to minimize ISS and found: the shallower the better. GF150/50 is very good in terms of ISS but will almost certainly cause a hit.
Nobody has ever stated that ISS was the sole factor in determining a profile's effectiveness. As Dr. Doolette has stated, "The integral supersaturation is an “algorithm-independent” way to look at the magnitude and duration of supersaturation of a schedule prescribed by any algorithm, it is unrelated to how that schedule was prescribed."

We have other reasons to believe that ignoring stops altogether, the shallowest profile, or your variant of using a low GF of 150, are not good decompression methods. It's critical to remember that in the context of the chart in question, we KNOW the DCS rates for two of the profiles -- the NEDU deep stops profile (A2) had a DCS rate of about 5% and the NEDU shallow stops profile (A1) had a rate of about 1.6%. The NEDU scientists believed that the difference in overall supersaturation exposure between the two profiles was the best explanation of the difference in DCS rates. So ISS is one interesting metric to review.
 
...
I then tried to minimize ISS and found: the shallower the better. GF150/50 is very good in terms of ISS but will almost certainly cause a hit.
...

Without checking the numbers, I'd expect at GF 150/50 all the stops to be controlled by GF High. Is that not what's happening? I.e. the results from 150/50 are different from (edit: a no-stop dive at) 50/50: 150 should generate stops above the surface and thus never kick in in practice so M0 is the only one that applies.
 
https://www.shearwater.com/products/perdix-ai/

Back
Top Bottom