A flexible, pervasive IoT ecosystem monitoring physiological and environmental parameters with real-time physician control — extending the work of Haghi et al.
Gap Analysis
Paper Limitations vs Our Innovation
Original Paper Limitations
No real-time physician control — sensors run continuously with no remote activation/deactivation
Fixed gateway / vendor lock-in — architecture tied to Ubiqsense + Equivital hardware only
No remote sensor activation — patients cannot receive updated sensing profiles from clinicians
Missing alert simulation — no threshold-based warning or notification infrastructure demonstrated
Unidirectional communication only — data flows sensor→cloud, no reverse command channel
Our Innovations
Physician-controlled sensor activation — clinicians toggle individual sensors on/off remotely in real time
Flexible multi-vendor gateway — supports Fitbit, Garmin, or any BLE wearable via abstract gateway layer
Real-time bidirectional communication — WebSocket channel enables cloud-to-device commands
Built as part of Reimagine Research With AI: Paper to Prototype — a Bayes Lab initiative transforming academic research into working demos.
🙏
Thank you, Bayes Lab!
We're grateful to Bayes Lab for giving us the opportunity to participate in
"Reimagine Research With AI: Paper to Prototype" — an inspiring challenge that pushed us to turn
a published IEEE research paper into a fully functional interactive platform.
Analysed the Haghi et al. paper, mapped the sensor architecture, and implemented the live data simulation with realistic physiological and environmental parameter ranges.
This prototype is based on the peer-reviewed paper by
M. Haghi, K. Thurow & R. Stoll —
"A Flexible and Pervasive IoT Based Healthcare Platform for Physiological and Environmental Parameters Monitoring",
IEEE Internet of Things Journal, Vol. 7, Issue 6, Dec 2020.
DOI: 10.1109/JIOT.2020.2982416.
Funded by the Ministry of Economics, Mecklenburg-Vorpommern (Project MV: TBI-V-1-127-VBW-044).
System Design
Technical Architecture
Three-tier IoT architecture enabling seamless data flow from body-worn sensors through a flexible gateway to the Azure p²Health cloud platform.
1
🩺
Sensor Layer
Ubiqsense wristband + Equivital EQ02 chest belt for physiological data. Environmental micro-sensors for gas, noise, UV, and temperature. Fitbit/Garmin as alternate wearable options.
UbiqsenseEquivitalFitbitBLE 5.0
2
📱
Gateway Layer
Smartphone-based gateway aggregates BLE sensor data. Performs local pre-processing, buffering, and AES-128 encryption before forwarding via HTTPS to cloud. Supports multi-vendor wearables.
BLE → HTTPSAES-128Multi-vendor
3
☁️
Cloud Layer
Azure-hosted p²Health microservices platform. Stores and processes time-series sensor data, provides physician dashboard, alert engine, and bidirectional WebSocket command channel.
AzureOAuth 2.0WebSocket
Monitored Parameters
Category
Parameter
Sensor
Typical Range
Alert Threshold
Physiological
Heart Rate
Ubiqsense / Fitbit
60–100 bpm
>120 bpm
Physiological
Respiration Rate
Equivital EQ02
12–20 brpm
>25 brpm
Physiological
RR Interval (HRV)
Equivital EQ02
600–1000 ms
<400 ms
Physiological
Skin Temperature
Ubiqsense
34–38°C
>38.5°C
Physiological
Sleep Stages
Ubiqsense
Light/Deep/REM
<4h total
Environmental
NO₂
Ubiqsense Env.
0–0.5 ppm
>1.0 ppm
Environmental
CO
MiCS-4514
0–9 ppm
>35 ppm
Environmental
O₃
MiCS-4514
0–0.07 ppm
>0.1 ppm
Environmental
Noise Level
MEMS Mic
30–65 dB
>85 dB
Environmental
UV Index
VEML6070
0–5
>6
Environmental
Ambient Temp
BME280
18–28°C
>35°C
Environmental
Humidity
BME280
30–70%
>85%
Environmental
Pressure
BME280
970–1030 hPa
<960 hPa
Security & Privacy
AES-128 Encryption
HTTPS / TLS 1.3
OAuth 2.0 Auth
Anonymized Data
GDPR Compliant
celisca Laboratory Results
Research Validation
Experimental results from the Center for Life Science Automation, University of Rostock, demonstrating platform performance across multiple scenarios.
NO₂ Spike Detection
Chemical reaction test: 0 → 21.8 ppm in 18 min
HR vs Respiration Sync
Activity correlation: 74–93 bpm / 7–21 brpm
7-Hour Sleep Monitoring
Ambient noise + HRV + sleep stage tracking
<5s
NO₂ Spike Detection Latency
7h+
Continuous Sleep Monitoring
13
Parameters Monitored Simultaneously
2-way
Bidirectional Communication Active
"Validated at Center for Life Science Automation (celisca), University of Rostock — real-time bidirectional communication established between physician dashboard and patient wearable gateway, with NO₂ spike detection confirmed within 5 seconds of chemical exposure."
Simulated Real-Time Stream
Live Demo Dashboard
Simulated live sensor feed auto-refreshing every 3 seconds. Values fluctuate realistically to demonstrate the monitoring platform.
Auto-refreshing every 3 seconds— just updated
Physiological
72
Heart Rate
bpm
14
Respiration
brpm
36.1
Skin Temp
°C
850
RR Interval
ms
Environmental
0.12
NO₂
ppm
38
Noise
dB
3
UV Index
/ 11
55
Humidity
%
Alert Simulation
Trigger a simulated high NO₂ threshold breach to demonstrate the real-time alert notification system.