amsys 19(18): e4

Research Article

Hypoglycemic Detection by Human Breath: A Mobile Health App that Alerts Diabetics of Low Blood Glucose

Download688 downloads
  • @ARTICLE{10.4108/eai.23-3-2018.162220,
        author={A. Faiola and H. Vatani and M. Agarwal},
        title={Hypoglycemic Detection by Human Breath: A Mobile Health App that Alerts Diabetics of Low Blood Glucose},
        journal={EAI Endorsed Transactions on Ambient Systems},
        volume={6},
        number={18},
        publisher={EAI},
        journal_a={AMSYS},
        year={2019},
        month={12},
        keywords={Hypoglycemia, diabetes, breathing sensor, mobile health, interface design, data visualization},
        doi={10.4108/eai.23-3-2018.162220}
    }
    
  • A. Faiola
    H. Vatani
    M. Agarwal
    Year: 2019
    Hypoglycemic Detection by Human Breath: A Mobile Health App that Alerts Diabetics of Low Blood Glucose
    AMSYS
    EAI
    DOI: 10.4108/eai.23-3-2018.162220
A. Faiola1,*, H. Vatani1, M. Agarwal2
  • 1: Dept. of Biomedical & Health Information Sciences, University of Illinois at Chicago, Chicago, IL, USA
  • 2: Dept. of Mechanical Engineering, Purdue University, IUPUI, Indianapolis, IN, USA
*Contact email: faiola@uic.edu

Abstract

Low blood glucose (BG) or hypoglycemia (HYPO) can lead to severe health complications such as weakness and unconsciousness. To avoid problems BG self-management is needed. We developed a non-invasive breathing system (HYPOalert) to detect HYPO in human-breath, that sends warning alerts and data visualization to monitor progress. This paper presents two HYPOalert prototype iterations with testing results. Of 14 Type 1/2 diabetics tested, only 10% were pleased with existing monitoring systems and 85% expressed interest in using HYPOalert more than 20x a day. The usability study showed that 92% agreed-strongly agreed with the HYPOalert design, including color/menus/navigation/typography; and 64% felt positive about the apps consistency, flexibility, and info architecture. A post-test survey provided a satisfaction score: 6.64/10, with an open-ended interview showing that HYPOalert could positively impact lifestyle practices, self-managing, and help advance an understanding of the disease.