Validation of a wearable biosensor device for vital sign monitoring in septic emergency department patients in Rwanda

Garbern, Stephanie Chow; Mbanjumucyo, Gabin; Umuhoza, Christian; Sharma, Vinay; Mackey, James; Tang, Oliver Y.; Martin, Kyle Denison; Twagirumukiza, Francois Regis; Rosman, Samantha L.; McCall, Natalie; Wegerich, Stephan; Levine, Adam C.

doi:10.1177/2055207619879349

Cited by 17 publications

(21 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the generalizability of the results can definitely be improved by increasing the sample size of the patient cohort in our study, to the best of our knowledge, this is the first prospective study in an LMIC setting to collect a relatively large dataset of high-resolution ECG signals for long time duration (24 h) from sepsis patients, and from this data propose ML-based prediction algorithms for a better care management of these patients. Another study [ 19 ] with a similar data collection protocol of sepsis patients in Rwanda performed statistical analysis with data from 43 patients. This attests to the fact that data collection for research purposes from critically ill patients from a critical unit for research purposes is a very cumbersome task, particularly in low-resource settings.…”

Section: Discussionmentioning

confidence: 99%

“…A study undertaken in Rwanda has shown the feasibility and high accuracy of wearable biosensor devices in monitoring vital signs of acutely ill paediatric and adult emergency department (ED) patients with sepsis in LMIC settings [ 19 ]. The authors demonstrated that vital sign measurements from a wireless wearable device are reliable and accurate compared to those obtained by an experienced nurse.…”

Section: Related Workmentioning

confidence: 99%

“…Furthermore, the integration of automatic prediction algorithms means wearable systems could be especially attractive in settings with limited staff. Despite the great potential of wearable sensors in resource-limited healthcare systems, very few studies have conducted research on the use of wearable systems in these settings for critically ill patients [ 18 , 19 ].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Sepsis Mortality Prediction Using Wearable Monitoring in Low–Middle Income Countries

Ghiasi

Zhu

Lü

et al. 2022

Sensors

View full text Add to dashboard Cite

Sepsis is associated with high mortality—particularly in low–middle income countries (LMICs). Critical care management of sepsis is challenging in LMICs due to the lack of care providers and the high cost of bedside monitors. Recent advances in wearable sensor technology and machine learning (ML) models in healthcare promise to deliver new ways of digital monitoring integrated with automated decision systems to reduce the mortality risk in sepsis. In this study, firstly, we aim to assess the feasibility of using wearable sensors instead of traditional bedside monitors in the sepsis care management of hospital admitted patients, and secondly, to introduce automated prediction models for the mortality prediction of sepsis patients. To this end, we continuously monitored 50 sepsis patients for nearly 24 h after their admission to the Hospital for Tropical Diseases in Vietnam. We then compared the performance and interepretability of state-of-the-art ML models for the task of mortality prediction of sepsis using the heart rate variability (HRV) signal from wearable sensors and vital signs from bedside monitors. Our results show that all ML models trained on wearable data outperformed ML models trained on data gathered from the bedside monitors for the task of mortality prediction with the highest performance (area under the precision recall curve = 0.83) achieved using time-varying features of HRV and recurrent neural networks. Our results demonstrate that the integration of automated ML prediction models with wearable technology is well suited for helping clinicians who manage sepsis patients in LMICs to reduce the mortality risk of sepsis.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Sepsis Mortality Prediction Using Wearable Monitoring in Low–Middle Income Countries

Ghiasi

Zhu

Lü

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…According to international guidelines of respiratory infection management, the RR levels are considered to vary in the order of tens of breaths per minute between normal and pathologically fast breathing, 1,10,24 which indicates that NAPPA's accuracy is clearly within the required limits. Substantial differences in individual RR levels are due to many more physiological factors, such as age, 1 altitude, 25 vigilance state, 16,26 position 27 or other acute medical conditions 28,29 . Respiratory assessment of subtle changes in such conditions may still need more stable and well‐established clinical methodology 10 .…”

Section: Discussionmentioning

confidence: 99%

An openly available wearable, a diaper cover, monitors infant's respiration and position during rest and sleep

et al. 2021

View full text Add to dashboard Cite

Surveillance of respiration, body posture and/or other physical activity are essential in a wide range of early childhood diagnostics and health monitoring. 1,2 Many advanced technologies exist for their tracking in laboratory or intensive care environments 3 ; however, they are challenging to assess in long-term surveillance, especially in out-of-hospital settings. While assessment of respiration rate (RR) would be crucial in situations like early detection of childhood pneumonia, 1,4,5 it has proven to be a challenge even when using human-assisted methods. [6][7][8][9][10] A viable solution for such surveillance should strike a balance between three interests: it should be sufficiently reliable and accurate,

show abstract

“…Miniaturized sensors attached or worn on the user's body, along with ambient sensors, are a new generation of computers that can be tailored to tackle healthcare management challenges. Applications that have been developed include chronic and non-communicable disease monitoring [1][2][3][4][5][6][7], fall monitoring [8][9][10], emergency care management [11][12][13], rehabilitation [14][15][16][17], fitness and lifestyle tracking [18,19], and sport performance monitoring [20][21][22].…”

Section: Introductionmentioning

confidence: 99%

A Dynamic Light-Weight Symmetric Encryption Algorithm for Secure Data Transmission via BLE Beacons

Banani

Thiemjarus

Wongthavarawat

et al. 2021

JSAN

View full text Add to dashboard Cite

Pervasive sensing with Body Sensor Networks (BSNs) is a promising technology for continuous health monitoring. Since the sensor nodes are resource-limited, on-node processing and advertisement of digested information via BLE beacon is a promising technique that can enable a node gateway to communicate with more sensor nodes and extend the sensor node’s lifetime before requiring recharging. This study proposes a Dynamic Light-weight Symmetric (DLS) encryption algorithm designed and developed to address the challenges in data protection and real-time secure data transmission via message advertisement. The algorithm uses a unique temporal encryption key to encrypt each transmitting packet with a simple function such as XOR. With small additional overhead on computational resources, DLS can significantly enhance security over existing baseline encryption algorithms. To evaluate its performance, the algorithm was utilized on beacon data encryption over advertising channels. The experiments demonstrated the use of the DLS encryption algorithm on top of various light-weight symmetric encryption algorithms (i.e., TEA, XTEA, PRESENT) and a MD5 hash function. The experimental results show that DLS can achieve acceptable results for avalanche effect, key sensitivity, and randomness in ciphertexts with a marginal increase in the resource usage. The proposed DLS encryption algorithm is suitable for implementation at the application layer, is light and energy efficient, reduces/removes the need for secret key exchange between sensor nodes and the server, is applicable to dynamic message size, and also protects against attacks such as known plaintext attack, brute-force attack, replaying attack, and differential attack.

show abstract

Validation of a wearable biosensor device for vital sign monitoring in septic emergency department patients in Rwanda

Cited by 17 publications

References 47 publications

Sepsis Mortality Prediction Using Wearable Monitoring in Low–Middle Income Countries

Sepsis Mortality Prediction Using Wearable Monitoring in Low–Middle Income Countries

An openly available wearable, a diaper cover, monitors infant's respiration and position during rest and sleep

A Dynamic Light-Weight Symmetric Encryption Algorithm for Secure Data Transmission via BLE Beacons

Contact Info

Product

Resources

About