Health Monitoring System Using Machine Learning Techniques Algorithm

Abstract

One of the most crucial aspects of someone\'s capacity to progress in life is their physical and mental well-being. Given the resources and requirements of society, the health-care system seeks to improve the populace as effectively as feasible. Due to a lack of timely medical equipment and treatments, death rates are growing in most nations. These health concerns can be prevented by offering standard healthcare services. The Flask framework was used to create the web application that houses our health monitoring system. In this Health Monitoring System, we employed Decision Tree Classification (Supervised Machine Learning method) to precisely anticipate outcomes. We used our own dataset to train and test our model. We could anticipate the patient\'s health level and area of risk based on that evaluation.

Introduction

I. INTRODUCTION

Proper health monitoring is the main problem of today. Patients experience severe health issues as a result of inadequate health monitoring systems. Today, a patient's health can be tracked online by a wide variety of gadgets. These tools are being fully utilised by medical practitioners to keep track of their patients' health. With the emergence of hundreds of new healthcare technology startups, machine learning is transforming the healthcare sector. In this study, we will develop a health monitoring system that keeps track of the patient's BMI, age, gender, body temperature, blood pressure, pulse rate, alcohol use, and smoking habits. This approach can assist people in managing a healthy lifestyle by providing early risk projections and suitable personalised advice. We propose to (a) identify health risk factors, (b) conduct data collection from controlled trials, (c) perform data analyses, and (d) perform a predictive analysis with machine learning algorithms for future health risk predictions and behavioural interventions in order to develop a system that is intelligent, automated, personalised, contextual, and behavioural recommendations to achieve personal wellness goals. This system employs the decision tree classification method, which contributes to high accuracy and reliable patient health risk level prediction.

The main issue today is proper health monitoring. Patients experience major health-related problems as a result of inadequate health monitoring systems. There are numerous tools available today for online patient health monitoring. Health professionals are fully utilising these tools to monitor the wellbeing of their patients. Machine learning is transforming the healthcare sector with the emergence of hundreds of new healthcare technology firms. We'll create a health monitoring system in this article that keeps track of the patient's BMI, age, gender, body temperature, blood pressure, pulse rate, if they drink alcohol, and whether they smoke. With accurate customised suggestions and early risk projections, this system can help people manage a healthy lifestyle.

II. LITERATURE REVIEW

1. Kartikee Uplenchwar et al a Raspberry Pi and Arduino-based IOT-based health monitoring system was developed. The transmitting part, the processing unit, and the receiving section make up the major three stages of this system. The biological sensors that make up the transmitting end are primarily employed to detect the bio potential signals coming from the patient's body [5]. By using wearable sensors, a set of five parameters—electrocardiogram (ECG), pulse rate, weight, temperature, and position detection—have been discovered. These sensors are linked to a Raspberry Pi and an Arduino. When the Raspberry Pi is online, it functions as a server and transmits data to a certain URL. On any mobile device, including connected computers and smartphones, the crucial parameters can be seen and monitored under same network. But this system has no live monitoring and data storing facility.
2. Stephanie B. Baker et al Internet of Things for Smart Healthcare: Technologies, Challenges, and Opportunities was the topic of a paper that was presented. The systems aimed at a particular condition could also use the specialised sensors, such as blood-glucose, fall detection, and joint angle sensors. Data is sent to the central node from sensor nodes attached to it. It processes the data that aids in the execution of some decisions before sending the data to a remote place. Machine learning algorithms have the ability to spot previously undetected trends in medical data, propose diagnosis and treatment plans, and provide advice to medical practitioners about specific patients. As a result, cloud storage systems ought to be created in a way that makes machine learning on large data sets possible [6].
3. Kirankumar et al used the Raspberry Pi 2 to construct a low-cost web-based human health monitoring system. This includes the patient's vital signs, including their body temperature, heart rate, and blood pressure. It also includes an alcohol sensor to check whether the patient has consumed alcohol, an ECG sensor, a sound sensor, an EMG sensor to check their level of stress, and a camera to record their live streaming video [7].The Raspberry Pi 2 microcontroller assists in gathering all these parameters, which are then displayed on a computer's Putty SSL Client. The Raspberry Pi's Wi-Fi module links the module to the internet by utilising the local Wi-Fi network. This makes it possible to keep track of a patient or a baby online thanks to a specially created webpage for it. The limitation of the system is that Cloud cannot identify specific doctor for consultation from the sensor data.

III. EXISTING SYSTEM

1. In the conventional system, the patient must receive a specific treatment in order to be cured; otherwise, his condition may worsen and he may even pass away. Unfortunately, the present monitoring systems frequently produce erroneous reports.
2. In actuality, the monitoring system may sound an alarm even when there is no genuine severe condition. However, in other instances, they are brought on by incorrect parameter or monitoring device settings.
3. In addition, the monitoring systems do not take into account how the observed parameters relate to one another. Each parameter is measured independently, which may produce inaccurate results. Therefore, fraudulent reports pose a serious risk to the patient's life.
4. They fail to report patients' actual conditions, which can complicate monitoring task more complicate. Furthermore, the working condition of the medical staff become more difficult and make patients under more pressure.

IV. PROPOSED SYSTEM

The main issue today is proper health monitoring. Patients experience major health-related problems as a result of inadequate health monitoring systems. There are numerous tools available today for online patient health monitoring. Health professionals are fully utilising these tools to monitor the wellbeing of their patients. Machine learning is transforming the healthcare sector with the emergence of hundreds of new healthcare technology firms. We'll create a health monitoring system in this article that keeps track of the patient's BMI, age, gender, body temperature, blood pressure, pulse rate, if they drink alcohol, and whether they smoke. With accurate customised suggestions and early risk projections, this system can help people manage a healthy lifestyle.

We propose to (a) identify health risk factors, (b) conduct data collection from controlled trials, (c) perform data analyses, and (d) perform a predictive analysis with machine learning algorithms for future health risk predictions and behavioural interventions in order to develop a system that is intelligent, automated, personalised, contextual, and behavioural recommendations to achieve personal wellness goals. This system employs the decision tree classification method, which contributes to high accuracy and reliable patient health risk level prediction.

The literature evaluation of heart disease prediction systems and overviews of current approaches are shown in the research article, which allows us to refine our method. In our method, we analysed various machine learning classification methods to manually and on the web platform predict the heart illness of each patient using the heart patients dataset from Alim et al. (2020).

The investigation reveals that Decision Tree and Random Forest approaches are 99% more accurate than other algorithms. Between these approaches, Decision Tree (0.0121) has a less classification error than Random Forest (0.0146).  The goal of this research's further expansion is to use more advanced machine learning approaches to diagnose heart problems with 100% accuracy. By creating an Android app, research will be expanded upon for bettering user accessibility.

VIII. ADVANTAGES

1. Applications of health monitoring using machine learning include early identification of cardiovascular diseases and cardiac disorders, as well as Clinical Decision Support System (CDSS) that can help doctors, nurses, patients, and other carers in making better decisions.
2. The system can also be used by regular individuals to identify whether they have a major health issue and to seek care by getting in touch with neighbouring hospitals.
3. Hundreds of new healthcare technology businesses are transforming the healthcare sector as a result of machine learning..

IX. DIS-ADVANTAGES

1. System is complicated.
2. System is Costly

X. APPLICATION

Applications of health monitoring utilising machine learning include early identification of cardiovascular diseases and chronic diseases, as well as Clinical Decision Support System (CDSS) that can help doctors, nurses, patients, and other carers in making better decisions. By contacting neighbouring hospitals, regular people can use this system to find out if they are suffering from a significant health issue and get care. With hundreds of new healthcare technology startups, machine learning is changing how the healthcare sector operates.

Conclusion

Instead of seeking treatment after a diagnosis, this research offers a way to prevent the condition through early intervention. With the help of the suggested system, healthcare workers can make better decisions, spot trends and innovations, and increase the effectiveness of research and clinical trials. It is also feasible to anticipate diseases more accurately. It enhances the way healthcare services are delivered, reduces costs, and carefully manages patient data.

References

[1] Jorge Gómez, \"Patient Monitoring System Based on Internet of Things: A survey\", Computer Networks, Vol.54, pp. 2787-2805, 2016. [2] Sneha N. Malokar , Samadhan D. Mali, \"Patient Monitoring System Based on Internet of Things using Rasperripy: Key features, application and open issues\", Computer Communication, Vol.54, pp., 2016. [3] Pioggia, \"Personal Health System architecture for stress monitoring and support to clinical decisions\", Computer Communications Vol.35, pp.1296-1305, 2017. [4] Franca Delmastro, \"Pervasive communications in healthcare\", Computer Communications, Vol.35, pp.1284-1295,2017. [5] KartikeeUplenchwar, Aditi Vedalankar “IoT Based Health Monitoring System using Raspberry Pi and Arduino” , International Journal of Innovative Research in Computer and Communication Engineering, Vol. 5, Issue 12, December 2017. [6] Stephanie B. Baker, Wei Xiang and Ian Atkinson, “Internet Of Things For Smart Healthcare: Technologies, Challenges, And Opportunities” IEEE transactions, Volume 5, 2017 . [7] Kirankumar and . Prabhakaran, “Design And Implementation Of Low Cost Web Based Human Health Monitoring System Using Raspberry Pi 2” , International Conference on Electrical, Instrumentation and Communication Engineering ,2017. [8] Rohan Bhardwaj, Ankita R. Nambiar , Debojyoti Dutta, “A Study of Machine Learning in Healthcare” IEEE Annual Computer Software and Applications Conference ,2017. [9] Gurjar, Neha and Sarnaik , “Heart Attack Detection By Heartbeat Sensing using Internet Of Things: IOT ” International Research Journal of Engineering and Technology , Volume: 05 ,Issue: 03 ,2018. [10] Pradnyavant Kalamkar, Pooja Patil, TriveniBhongale, MeghaKamble, “Human Health monitoring System using IOT and Raspberry pi3” International Research Journal of Engineering and Technology , Volume 5 , No 3 , 2018 .

Copyright

Copyright © 2023 Rutika Bhagat, Prof. Pragati Patil. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.