Andreas Fichtner, Björn Hannesen, Felix Stein, Benedikt Schrofner-Brunner, Thomas Pohl, Thomas Grab, Thea Koch, Tobias Fieback
{"title":"Predicting Post-Dive Inert Gas Bubble Grades in Non-Decompression Scuba Diving with Air: Simplified Model for Enhanced Diver Safety.","authors":"Andreas Fichtner, Björn Hannesen, Felix Stein, Benedikt Schrofner-Brunner, Thomas Pohl, Thomas Grab, Thea Koch, Tobias Fieback","doi":"10.1186/s40798-025-00832-x","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Even well-planned no-decompression dives can still produce inert gas bubbles that increase decompression sickness risk. A previously proposed formula for predicting post-dive bubble grades integrates individual factors (age, breathing gas consumption) with dive parameters (maximum depth, surface interval). This study aimed to confirm the formula's validity in an independent dataset and to find out whether detailed dive profile data are of further relevance in predicting echocardiography-derived post-dive bubble grades. Additionally, we explored whether machine learning models leveraging detailed dive profile data could enhance predictive accuracy.</p><p><strong>Results: </strong>A total of 59 divers performed 359 no-decompression open-circuit air dives in freshwater and saltwater. Post-dive transthoracic echocardiography detected bubbles (Eftedal-Brubakk grade ≥ 1) in 29.8% of dives. Maximum depth, total dive time, air consumption, and age correlated significantly with observed bubble grades (r<sub>s</sub>=0.37, r<sub>s</sub>=0.16, r<sub>s</sub>=0.27, r<sub>s</sub>=0.13, respectively). The original prediction formula remained valid (r<sub>s</sub>=0.39) and adequately captured higher-grade dives. Spending additional time in shallow water after deep segments reduced bubble formation. Machine learning approaches based on typical dive computer data (e.g. dive profile) provided stronger predictions (r<sub>s</sub>=0.49).</p><p><strong>Conclusions: </strong>This study shows that maximum depth, age, surface interval and total breathing gas consumption are sufficient predictors of post-dive bubble load in no-decompression air dives. This allows divers to individually adopt bubble-reducing measures-such as resting, hydrating, and extending surface intervals-once alerted to a higher-risk class. Integrating the formula into dive computers may offer real-time, individualised risk guidance and help prevent decompression sickness despite following computer-derived profiles in recreational diving.</p>","PeriodicalId":21788,"journal":{"name":"Sports Medicine - Open","volume":"11 1","pages":"29"},"PeriodicalIF":4.1000,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11953488/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Sports Medicine - Open","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1186/s40798-025-00832-x","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"SPORT SCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Even well-planned no-decompression dives can still produce inert gas bubbles that increase decompression sickness risk. A previously proposed formula for predicting post-dive bubble grades integrates individual factors (age, breathing gas consumption) with dive parameters (maximum depth, surface interval). This study aimed to confirm the formula's validity in an independent dataset and to find out whether detailed dive profile data are of further relevance in predicting echocardiography-derived post-dive bubble grades. Additionally, we explored whether machine learning models leveraging detailed dive profile data could enhance predictive accuracy.
Results: A total of 59 divers performed 359 no-decompression open-circuit air dives in freshwater and saltwater. Post-dive transthoracic echocardiography detected bubbles (Eftedal-Brubakk grade ≥ 1) in 29.8% of dives. Maximum depth, total dive time, air consumption, and age correlated significantly with observed bubble grades (rs=0.37, rs=0.16, rs=0.27, rs=0.13, respectively). The original prediction formula remained valid (rs=0.39) and adequately captured higher-grade dives. Spending additional time in shallow water after deep segments reduced bubble formation. Machine learning approaches based on typical dive computer data (e.g. dive profile) provided stronger predictions (rs=0.49).
Conclusions: This study shows that maximum depth, age, surface interval and total breathing gas consumption are sufficient predictors of post-dive bubble load in no-decompression air dives. This allows divers to individually adopt bubble-reducing measures-such as resting, hydrating, and extending surface intervals-once alerted to a higher-risk class. Integrating the formula into dive computers may offer real-time, individualised risk guidance and help prevent decompression sickness despite following computer-derived profiles in recreational diving.