SHUBHCHINTAK: An efficient remote health monitoring approach for elderly people

Abstract

With the proliferation of IoT technology, it is anticipated that healthcare services, particularly for the elderly persons, will become a major thrust area of research in the coming days. Aim of this work is to design a fit-band containing multiple sensors to provide remote healthcare services for the elderly persons. An application has been designed to capture health data from the fit-band, pre-process the data and then send them to cloud for further analysis. A wireless Bluetooth enabled connection is proposed to establish communications between sensors and the application for data transmission. In the proposed application, there are three different front-end interfaces for three different users: system administrator, patient and doctor. The data collected from the patient's fit-band are sent to a cloud data storage, where the data will be analyzed to detect anomaly (e.g., heart attack, sleep apnea, etc.). A Convolution Neural Network (CNN) model is proposed for anomaly detection. For the classification of anomaly, a Long Short Term Memory (LSTM) model is proposed. In the presence of anomaly, the system immediately connects a doctor through a phone call. A prototype system termed as Shubhchintak has been developed in Android/IOS environment and tested with a number of users. The fit-band provides data tracking with an overall accuracy of 99%; the system provides a response with 3000 requests in less than 100 ms. Also, Shubhchintak provides a real-time feedback with an accuracy of 97%. Shubhchintak is also tested by patients and doctors of a nearby hospital. Shubhchintak is shown to be a simple to use, cost effective, comfortable, and efficient system compared to the existing state of the art solutions.

Keywords: Artificial intelligence approach to health data classification; Cloud-based application; Health anomaly for elderly people; Remote health monitoring; Wireless sensors.

Fig. 1

Features of Shubhchintak healthcare application system

Fig. 2

A system-level architecture of Shubhchintak

Fig. 3

Brief Overview of sensors (a) accelerometer, (b) gyroscope, (c) ECG sensor, (d) pulse sensor, (e) blood oxygen level detector, (f) human body temperature sensor

Fig. 4

A brief overview of the circuit diagram to design the fitness band

Fig. 5

Anomaly detection and feedback mechanism in case of (a) non-prime users and (b) prime users

Fig. 6

Proposed CNN architecture with filter and kernel for each convolutional and pooling layer

Fig. 7

Overview of NLP model

Fig. 8

(a) Login page (b) Main page (c) Chatbot assistant (d) Geo-Sensing (e) Health data (Heart Rate and Step Count) and (f)Auto calling screen of the Shubhchintak application

Fig. 9

Health data of the shubhchintak application. (a)Heart Rate,(b) Step Count (c) Blood Oxygen Level, (d) Basic Metabolism Rate, (e) Monthly Health Condition (f) Temperature

Fig. 10

True positive rate vs false positive rate and precision-recall curve for anomaly detection model

Fig. 11

Classification accuracy and cross-entropy loss of anomaly detection model

Fig. 12

Performance analysis of the chatbot

Fig. 13

Comparison analysis based on request-response time

Fig. 14

Comparison analysis based on the number of messages per latency

Fig. 15

Comparison of the scalability and pricing with state-of-the-art-approaches

Fig. 16

Comparison of data storage cost