SensoMed: Cloud Based Remote Medical Sensing Solution

Overview

Residents of rural and remote areas do not always have the best access to a doctor, requiring them to make a long trip to the hospital. This is especially inconvenient for patients with motor difficulties, chronic illnesses or those who are recovering from a recent surgery. A remote sensing device to measure standard vitals such as blood pressure, respiration rate, SpO2,­ body temperature and heart rate, was developed to log data to the cloud through a smart phone.

Monitoring these parameters remotely gives doctors a way to diagnose potentially dangerous conditions without causing inconvenience to the patient. Potential users of the product could be frequent travelers, elderly people, patients in remote locations outside the city, immobile and bedridden patients, infants and patients suffering from chronic illnesses that need frequent monitoring.

Such a device could help doctors schedule an early discharge, saving patients a lot of money. Along with patients, the device would also help doctors through early diagnosis, analysis and log maintenance of patient data, integration into existing mobile and web platforms and integration of pharmacy records for accurate with history maintenance. Being a medical application, the solution could also run data analysis on the patient logs to automate some portion of the diagnosis, analyze patterns in an area and provide awareness on ongoing medical programs to the patient.

How it Works

The device needs to be able to monitor patient parameters remotely, be easy to use, low power and support secure data transfer. Figure 1 provides an overview of the solution from the patient to the doctor. An embedded device was developed that could read SpO2, heart rate and temperature from sensors into a micro controller and transfer the data to a phone application through Bluetooth.

While NFC is a lower power solution which could also potentially power the device, BLE was selected due to the range required for a wearable product that was intended in the design. The phone application would then periodically push the data to the cloud through an existing Internet connection. If no Internet connection is available and the data points to a critical situation, an additional feature of sending alarms to a web server through SMS could be implemented with a service like Twilio. This data could then be serviced through a web application or a phone application by the doctor to monitor the patient.

The MAX30101 sensor

The MAX30101 sensor was used to measure pulse oximetry and calculate SpO2 concentrations, heart rate. The sensor includes a red, green LED transmitter and a photo receiver to estimate absorption of these wavelengths by the blood beneath the skin of the finger or wrist which can be then used to measure SpO2 levels. The frequency of the peaks in this measurement can be used to derive heart rate.

Theoretically, the sensors can achieve 1.5% accuracy for SpO2 and 5% accuracy on the heart rate. An existing library from the mbed repository was modified and used to derive the required data from the sensors. The algorithm used here can be further optimized to achieve better accuracy and reliable data.

The MAX32625 micro controller (MCU), features an M4F ARM cortex core with I2C and analog peripherals was used on the embedded device to read the sensors through I2C. The mbed Web IDE was used to program the MCU on the evaluation board from MAXIM. The DS3231 was used for temperature measurement with a ready to use library. Both sensors were connected to the MCU through I2C.

The NexPaq Platform

To implement communication with the phone, the nexpaq platform was leveraged which abstracts much of the lower level BLE communication with the MCU from the phone and provides a simpler Web API (CSS, HTML and Javascript) handle for application development. The calculated values of heart rate, SpO2 and temperature were parsed as required by the nexpaq application and sent to the phone.

An Android application tile was developed through customization of existing Android application source codes on the nexpaq platform. The application provides an on screen display of the measured parameters and pushes the data periodically to the cloud by raising an XML HTTP post request. The platform’s cloud API was used to move the data to the existing nexpaq cloud database.

The parameter data was then logged in a simple browser by raising a GET request. A web app can be developed in the future to make the display of the data intuitive to the medical professionals analyzing the log of patient records.

Conclusions

To conclude, a remote monitoring device for medical applications was developed within 48 hours by leveraging MAXIM and nexpaq platforms. With refinements in the sensing algorithm and application development, the product can be used to service patients in remote areas by logging their vitals directly to a platform accessible by their doctor.