Journal article Open Access

Design and Validation of a Breathing Detection System for Scuba Divers

Altepe, Corentin; Egi S. Murat; Ozyigit, Tamer; Sinoplu, D. Ruzgar; Marroni, Alessandro; Pierleoni, Paola


Citation Style Language JSON Export

{
  "note": "This paper is published as Open Access in the journal Sensors 2017, 17(6), 1349 and can be downloaded from the journal's webpage here: http://www.mdpi.com/1424-8220/17/6/1349. This paper has received funding from the European Union (EU)'s H2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 643712 to the project Green Bubbles RISE for sustainable diving (Green Bubbles). This paper reflects only the authors' view. The Research Executive Agency is not responsible for any use that may be made of the information it contains. \nThis is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. (CC BY 4.0).", 
  "DOI": "10.3390/s17061349", 
  "container_title": "Sensors", 
  "title": "Design and Validation of a Breathing Detection System for Scuba Divers", 
  "issued": {
    "date-parts": [
      [
        2017, 
        6, 
        9
      ]
    ]
  }, 
  "abstract": "<p><strong>Abstract</strong></p>\n\n<p>Drowning is the major cause of death in self-contained underwater breathing apparatus (SCUBA) diving. This study proposes an embedded system with a live and light-weight algorithm which detects the breathing of divers through the analysis of the intermediate pressure (IP) signal of the SCUBA regulator. A system composed mainly of two pressure sensors and a low-power microcontroller was designed and programmed to record the pressure sensors signals and provide alarms in absence of breathing. An algorithm was developed to analyze the signals and identify inhalation events of the diver. A waterproof case was built to accommodate the system and was tested up to a depth of 25 m in a pressure chamber. To validate the system in the real environment, a series of dives with two different types of workload requiring different ranges of breathing frequencies were planned. Eight professional SCUBA divers volunteered to dive with the system to collect their IP data in order to participate to validation trials. The subjects underwent two dives, each of 52 min on average and a maximum depth of 7 m. The algorithm was optimized for the collected dataset and proved a sensitivity of inhalation detection of 97.5% and a total number of 275 false positives (FP) over a total recording time of 13.9 h. The detection algorithm presents a maximum delay of 5.2 s and requires only 800 bytes of random-access memory (RAM). The results were compared against the analysis of video records of the dives by two blinded observers and proved a sensitivity of 97.6% on the data set. The design includes a buzzer to provide audible alarms to accompanying dive buddies which will be triggered in case of degraded health conditions such as near drowning (absence of breathing), hyperventilation (breathing frequency too high) and skip-breathing (breathing frequency too low) measured by the improper breathing frequency. The system also measures the IP at rest before the dive and indicates with flashing light-emitting diodes and audible alarm the regulator malfunctions due to high or low IP that may cause fatal accidents during the dive by preventing natural breathing. It is also planned to relay the alarm signal to underwater and surface rescue authorities by means of acoustic communication.</p>", 
  "author": [
    {
      "family": "Altepe, Corentin"
    }, 
    {
      "family": "Egi S. Murat"
    }, 
    {
      "family": "Ozyigit, Tamer"
    }, 
    {
      "family": "Sinoplu, D. Ruzgar"
    }, 
    {
      "family": "Marroni, Alessandro"
    }, 
    {
      "family": "Pierleoni, Paola"
    }
  ], 
  "page": "1349", 
  "volume": "17", 
  "type": "article-journal", 
  "issue": "6", 
  "id": "826188"
}
60
36
views
downloads
Views 60
Downloads 36
Data volume 102.5 MB
Unique views 59
Unique downloads 34

Share

Cite as