Machine Learning Driven IoT Based Smart Health Care Kit

Authors

Lekhasree Narayanagari
Department of Mathematical and Physical Sciences, Concordia University of Edmonton, Alberta, T5B 4E4, Canada
Baidya Nath Saha
Department of Mathematical and Physical Sciences, Concordia University of Edmonton, Alberta, T5B 4E4

Synopsis

This paper focuses on developing a machine learning driven IOT based smart healthcare kit. It plays an important role in emergency medical service like Intensive Care Units (ICU), by using an INTEL GALILEO 2ND generation development board. It facilitates to monitor and track different health indicators such as Blood Pressure, Pulses, and Temperature of the patient. This system allows to send the real time data of a patient to the physician and record it for future use. In this research we conducted two experiments: a)heart disease prediction from pathology data and b) lung disease prediction from X-ray images. For heart disease prediction we evaluate the performance of K-Nearest Neighbour and Random Forest Classifier and for lung disease prediction, we use VGG19 deep architecture. Experimental results demonstrate that machine learning can help to automate the IoT based smart healthcare kit and help doctors to diagnose the diseases.

ICTCon2021
Published
July 12, 2021
Online ISSN
2582-3922