mVitals An intelligent Edge Computing Based Wireless Mobile Healthcare System

Project Title

mVitals An intelligent Edge Computing Based Wireless Mobile Healthcare System

Project Area of Specialization Computer ScienceProject Summary

Internet of Things (IoT) allows effective and flexible real-time health care monitoring systems, equipped with sensors which collect the patient’s data regarding vital signs and reduce human error. The existing systems use cloud computing technology where the collected data is stored, processed and analyzed on cloud servers. The proposed system presents a personal healthcare system that will monitor the health parameters dynamically by making use of integrated wearable sensors with the incorporation of edge networking technology. It will use different sensors like ECG, heart rate sensor, body temperature sensor, blood pressure sensor and breathing rate sensor to monitor patient’s health. The smart system will use deep learning algorithms, which learn from past data of patients and provide predictions about the critical condition of patients. Complete filtration will be applied on acquired data to remove noise. In case of abnormal readings, the smart decision support system will generate alert messages for caretakers and medical staff. Medical history and reports of the patients will be stored in a real-time database which will further help doctors to assess the patient’s condition and suggest treatment plans based on deep learning.

Project Objectives

The main objective of the project are as follows:

• Design and implement an intelligent as well as a portable system for real-time healthcare monitoring.

• Develop a smart medical monitoring system, that will not only be utilized in hospitals, ambulances, etc. but also collaborate with the smart home idea to make it part of daily life activities.

• Deploy deep learning techniques for training and testing of models, to provide reliable disease diagnosis.

• Design a system incorporating a Clinical Decision Support System (CDSS), which will enhance the patient’s diagnosis with better analytics.

• Assist the doctors/caretakers to monitor the patient and provide a reliable notification mechanism in case of any criticality.

• Make use of wireless technology, to get rid of jumbled up wired hardware systems and allowing patients to roam freely.

• Develop an edge computing based system backed up with cloud services, to provide storage and analysis of data in real time.

• Store and manage the health record of patients for different purposes.

• Provide adequate visual representation of patients’ vital signs and ECG, so even a non-medical person can monitor the patient.

• Reduce the costs associated with employing expensive monitoring equipment.

Project Implementation Method

For the implementation of this project, Scrumban, an agile project development methodology that is hybrid of Scrum and Kanban, will be deployed. Scrumban is widely used in development and maintenance of projects. We are adapting this methodology because it embraces the features of both Scrum and Kanban where former is used as a way of working and latter is used to view, understand and improve performance. It also uses key metrics to estimate the average time for completion of a specific task, a highly versatile approach for workflow management, reduces the impact of errors, increases productivity and waste minimization efficiency.

Benefits of the Project

The proposed system will be mobile and wearable enabling wireless communication to continuously monitor vitals signs and ECG of a patient. The system will store records of patients' history as well as real-time monitoring data for future reference. The data will be stored in a secure and efficient database system and only authorized people will have the remote accessibility of a patient's medical record regardless of whereabouts with high processing speed. Incorporation of Clinical Decision Support System (CDSS), will enhance the patient’s diagnosis and provide better analytics using deep learning techniques. The system will aid in devising treatment plans for patients based on deep learning and past data. This cost-effective and efficient system will represent data in an easy to understand manner. Alerts and notifications will be sent to concerned personnel in case of any critical change in health parameters of patients so that the patient never stays unobserved.

Technical Details of Final Deliverable

The proposed system is an integration of hardware and software. Hardware will include a Wi-Fi router, the sensors for vital signs i.e. ADS1292R for breathing rate, CPS120 for blood pressure, LM-35 for body temperature, XD-58C for heart rate, ECG Module (AD8232), ECG Electrodes, ECG Electrode Connector, and a microcontroller called Arduino Nano (ATmega328P). Arduino Nano will be programmed using C++ programming language on Arduino IDE. All the sensors will be connected to the microcontroller, and wireless communication will enable the monitoring of patients’ real-time data. The information will be displayed on a web based system and a mobile application as well. The cross-platform data sharing will be done using RESTful APIs. The web based system will be developed using NodeJS for backend scripting and React for frontend. The mobile application will be developed using a framework for developing React apps called REACT Native and programming language such as Python or Java. Collected data of patients will be managed and stored on a cloud server. The deep learning algorithms and techniques will be applied on the patients' data to make the proposed system intelligent, using Python on an IDE called PyCharm. A secure connection, between the microcontroller and the system, will be established using Raspberry PI for edge network computing. Once the connection is established, the microcontroller will fetch patients’ data and transfer it to the network. After analyzing and processing the data, feature extraction will be performed for machine learning. Afterwards, the extracted features will be divided into test and train data. Deep learning techniques will be utilized to diagnose the patient based on monitoring data as well as considering the past learning of the system. It will also suggest treatment plans based on previous learning from recommendations of medical experts. Intelligent alert and notification mechanism will be implemented in case of any critical change in patient's health parameters so that the patient can receive the best possible observation and monitoring.

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther Industries Medical , Health Core Technology Wearables and ImplantablesOther Technologies Artificial Intelligence(AI), Internet of Things (IoT)Sustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
Raspberry PI 4	Equipment	2	17500	35000
Programming Boards	Equipment	4	3000	12000
Vital Sign Sensors	Equipment	24	700	16800
Connectors and Electrodes for ECG	Equipment	50	124	6200
Cloud Services, printing, stationary, overheads	Miscellaneous	1	10000	10000