Dopplering after diving – does it make sense?
For a while, at popular dive spots, you could often observe someone pulling a small portable ultrasound Doppler device out of their bag after a dive. Measuring bubbles. In the middle of the group, half curious, half proud. This happened to me once in Dahab, in a dive group where we – looking back, quite typically – primarily defined ourselves by depth.
Back then, I was still young, very keen on depth, and grateful for any form of validation. No bubbles, despite considerable depth – that felt like proof. Clearly, we had done something right during the ascent. Which was true: we spent a very long time shallow at the end, and even with my current knowledge, that is the crucial lever for getting out of the water as cleanly as possible. The fact that more bubbles were measured in the other group, which had only dived to around 30 meters, fit perfectly into the picture. Dive deep, ascend cleanly, everything done right. Mission accomplished.
Looking back today, this assessment seems quite naive to me. Not wrong in the strict sense, but far too simplistic. Back then, I read bubbles like a school grade: few bubbles equal good diving, more bubbles equal bad diving. What I didn’t see was everything in between – and everything that this simple equation didn’t explain.
Because bubbles and decompression sickness are related, but not in a simple way. Bubbles are a sign of decompression stress, not illness. Most bubbles measurable after diving remain asymptomatic, are tolerated, filtered, or reabsorbed by the body. Bubbles are neither proof of safety nor proof of danger. They are an indication – and a rather vague one at that.
This is precisely where the actual question of this article begins. Not: Can bubbles be measured? – yes, they can. But: What does that actually mean? And above all: What can be meaningfully deduced from it – and what cannot?
Measurable does not mean interpretable
The fact that bubbles are measurable is their greatest advantage. The fact that they are difficult to interpret is their biggest problem. The relationship between bubbles and symptoms remains unsatisfactorily vague to this day. Statistically, there are correlations: on average, the risk increases with higher bubble grades. But individually, the picture is so unstable that from a single measurement, one usually reads exactly what one already wanted to believe.
The chaos has a name: intra-personal variability. The same dive produces completely different bubble loads in different people. And even in one and the same person, the result for an identical profile can jump from “hardly anything” to “significant” without any visible change in gas, ascent, or behavior.
David Doolette took the trouble to compile data from several US Navy studies. Each point here is a bubble measurement, each line links the measurements for one person with identical profiles. And what can be seen here is very clear: for most people, the bubble load varies from day to day for one and the same profile. Bubble measurements are useful as a marker for decompression stress, but largely unsuitable as an individual predictive tool for “will I get DCS?”. And even less can one derive a prediction for the next dive from them.
O’Dive: technically convincing, scientifically challenging
Despite all doubts about bubble measurements, some divers wish to better assess the quality of their own Deco, and for this, in the absence of other markers, to measure bubbles. This is feasible with small, portable devices, but difficult. And this is where O’Dive comes in. Technically, the system is elegant: bubble measurement in the subclavian region, coupled with the dive profile, followed by algorithmic analysis and feedback. The measurement approach is plausible, the position is highly reproducible, and the whole thing is significantly more practical than most mini-ultrasound devices. The measurement is really simple, an app shows if it worked, and the analysis happens automatically. So the bubble load does not have to be interpreted by a human, which is always a pitfall.
Photo from Kreidesee Hemmoor
The detection of VGE in the subclavian vein area is well described in the literature: VGE can be reliably detected there, allowing for the assessment of decompression stress – as a measurement principle, it is solid, and the device works so well that it is now widely used by researchers, military, and professional divers.
The critical part begins where O’Dive goes beyond “measuring” and slips into “evaluating” and “recommending.” The app doesn’t just reveal the bubble load; it assesses it together with the dive profile that must be entered. This results in an evaluation of the Deco quality and a recommendation on what to change. But the algorithm behind it is not public. The assumptions, weightings, training data, and validations are not comprehensible from the outside. This is not a moral reproach, but a structural limitation: when a system derives a recommendation from data and an algorithm, transparency is the currency with which it buys trust. If this transparency is missing, all that remains is: “Trust me.”
So the device is great, the idea behind it really good – but precisely because it’s so unclear how bubbles should be assessed on an individual level, a mysterious “black box” that issues a recommendation is not for everyone.
Data: the greatest potential – and the shakiest point
However, there’s a thought here that is almost painful in this context due to its sheer obviousness: If the data collected there – standardized bubble measurements, cleanly linked to dive profiles, across many individuals and many repetitions – were available to independent research, that would be an enormous benefit. Precisely such datasets are often missing when we talk about individual variability, risk markers, and model validation, and they are extremely difficult to obtain.
Regarding the question of whether and how this data can be scientifically used, there are now contradictory statements. The device is used in studies, but a few years ago it was clearly stated that even the company had no access to the uploaded user data. At the same time, one hears from sellers that the model would get better and better through the data – but: does it? And how exactly?
We will probably have to wait and see. The potential exists – but currently it is not clear and transparent under what conditions it will actually be realized. Perhaps such devices and anonymous data collections hold one of the keys to understanding Deco stress significantly better in a few years than today – or perhaps chasing bubbles as an indicator is a dead end. We remain curious.
The three “bubble types”
Looking at the data on bubble variability, it’s clear that not everyone reacts differently every time – for some, measurements are consistent. Roughly three groups can be distinguished in the data.
There are divers who consistently develop few or no measurable bubbles. There are others who repeatedly show high bubble grades after comparable profiles. And then there is the large majority: individuals for whom one and the same profile sometimes leads to many, sometimes to hardly any measurable bubbles. It is precisely this intra-personal variability that makes individual risk assessments so difficult.
In research, it is always a pleasure to get “constant bubblers” as subjects. With them, a correlation between profile changes and bubble load can be observed at least somewhat stably. This is not true for the large group of “chaotic bubblers.”
Regular bubble measurements can indeed be useful – but not in the sense of evaluating individual dives. Their real value lies in finding out, over several measurements, which of these three groups one belongs to. Only then does a meaningful question about consequences arise. And these consequences look very different depending on the group.
What bubble measurements can – and cannot – do
What they can do
Bubble measurements provide one of the few directly measurable markers for Deco stress. They are comparatively easy to record and are well suited for comparing Deco procedures, profiles, or gas strategies – especially in research.
What they cannot do
They do not allow for reliable individual risk prediction for a single dive. The connection between bubbles and DCS is simply too unclear for that, and intra-personal variability also stands in the way.
The danger
Bubble measurements can create false security – or unnecessary fear. Both are problematic if measurements are considered in isolation and slip into decisions without understanding their limitations.
So: measure – yes. But for what?
If you want to use bubble measurements as a personal tool, it’s worth reversing the order. Don’t measure first and then reflexively interpret, but first recognize a pattern over several measurements and then consider what you want to do with it.
For people with consistently few bubbles, the benefit is often primarily psychological: confirmation that their way of diving is apparently well tolerated. For people with consistently high bubble loads, the benefit can be more practical: design gentler profiles, rethink gas strategies, change ascents – and observe whether the pattern shifts stably. And for the large group of “chaotic bubblers,” the benefit is often limited, because individual values are rarely more than an echo from a chaotic system.
Conclusion: Bubbles are an indication, not a judgment
Bubble measurements are not meaningless. On the contrary: as a marker for Deco stress, they remain a central tool in research, and they will probably remain the most tangible anchor we have for a long time.
But bubbles are not a judgment on a dive, and they are not a personal safety index. Those who measure them should treat them for what they are: an indication. A puzzle piece. And sometimes simply another reason to kindly distrust one’s own need for simple answers.
