Search results
(1 - 2 of 2)
- Title
- Implementation of a multisensor wearable artificial pancreas platform: ensuring safety with communication robustness and cyber security
- Creator
- Lazaro Martinez, Carmen Caterina
- Date
- 2019
- Description
-
Advances in IoT technologies and new sensor capabilities contributed to the rapid growth of wearable medical devices. Today, mobile sensor...
Show moreAdvances in IoT technologies and new sensor capabilities contributed to the rapid growth of wearable medical devices. Today, mobile sensor platforms can be effectively, cost efficiently integrated in healthcare applications. However, the increased risks of these devices, inherent vulnerabilities of mobile operating systems and open nature of the wireless protocols call for improved safety and security measures to prioritize patient's well-being. In the field of type 1 diabetes, blood glucose level management with insulin control algorithms are available in diabetes therapy systems, though none are fully automated and require extra announcements (such as meal and exercise) to operate. A mobile artificial pancreas (AP), based on Android smartphone, is developed: such a platform relies on off-the-shelf components and receives in real-time the physiological measurements from the wrist worn physical activity tracker and the glucose measurements, then used in a predictive control algorithm (originally developed and tested on a laptop), to compute the optimal amount of insulin to administer via an insulin pump. A dedicated remote server provides additional support for registration, authentication and data backup.The nature of the algorithm required a fast, reliable method to translate its inherent functions. Therefore, we implement a new semi-automatic conversion mechanism which ports MATLAB to Android as native C code. Validation tests of the mobile version confirm there are no deviations in the results.Moreover, in order to enhance safety guarantees for the patient, this cyber-physical system needs a robust implementation also resilient to attacks and failures. A central monitor module is introduced, wherein wireless devices and communications channels are integrated with complementary alarm and safety subsystems. The parameterization of the AP as a state machine demonstrates the efficiency to detect and react to possible errors, since any state change triggers the appropriate correcting response. The result is a protected and fail-safe environment, further expanded with security modules enforcing encryption, authenticated access and data-flow rules for intrusion detection.Overall, this research demonstrates, in the case of an AP, how challenges in diverse fields such as sensor fusion, control systems, wireless communications and cybersecurity can be addressed with a holistic approach for mobile health (mHealth).
Show less
- Title
- Wireless Body Sensor Network for Tracking Human Mobility using Long Short-Term Memory Neural Network for Classification
- Creator
- Gupta, Saumya
- Date
- 2019
- Description
-
A large number of sensors are used without justification of the number chosen or placement choice. Many papers about body sensor networks...
Show moreA large number of sensors are used without justification of the number chosen or placement choice. Many papers about body sensor networks explore how to capture a type or types of motion, but all their sensors are placed in different locations; making their algorithms very specific to that movement. In this research, we explore the enhancement of human activity classification algorithm using long short-term memory (LSTM) neural network and wearable sensor network. There are five identical nodes used in the body sensor network to collect data. Each node incorporates an ESP8266 Microcontroller with Wi-Fi which is connected to an inertial measurement unit consisting of triple axis accelerometer and gyroscope sensor board. An analysis on the accuracy that each sensor node provides separately and in different combinations has been conducted to allow future research to focus their positioning in optimal positions. A Robot Operating System (ROS) central node is used to illustrate the in-built multi-threading capability. For demonstration, the positions chosen are waist, ankles and wrists. The raw sensor data can be observed on screen while it is being labelled live to create fitting dataset for developing an artificial neural network. Expectation is that increasing the number of sensors should raise the overall accuracy of the output but that isn’t the case observed, positioning of the sensor is pertinent to improvement. These platforms can be further extended to understand different motions and different sensor positions, also expanded to include other sensors.
Show less