IOT based Patient Health Monitoring System using ML
Pramit Gupta1, Arinjay Bisht2
1Pramit Gupta, Pursuing Bachelors in Technology, Electronics and Communication Engineering ,Vellore Institute of Technology.
2Arinjay Bisht, Pursuing Bachelors in Technology, Electronics and Communication Engineering ,Vellore Institute of Technology.
Manuscript received on September 23, 2019. | Revised Manuscript received on October 15, 2019. | Manuscript published on October 30, 2019. | PP: 5086-5091 | Volume-9 Issue-1, October 2019 | Retrieval Number: A2148109119/2019©BEIESP | DOI: 10.35940/ijeat.A2148.109119
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The project focuses on the usage of sensing and analysis with the help of relevant sensor technologies in order to record the health conditions of people. The best way to understand this is with an example. A practising doctor who is not equipped with such technology can check the patients’ health only when the patient pays a visit to the clinic. Now, with the application of the proposed technological measures, the doctor would have a complete record of the patient whether at home, office or on the road, and this would enable him to prescribe medication in a much more efficient and effective manner. Also, it is important to appreciate that on the basis of patient data recorded in the past, a prediction model could help the doctor see irregularities and predict if a patient suffers from commonly occurring ailments hence saving time in an initial diagnosis. This method for Healthcare Data Analytics using Support Vector Machine (SVM) Algorithm helps improve accuracy when it comes to checking for specific diseases or figuring out the right treatment. Furthermore, the doctor can give a more personalized view if he has access to a large chunk of the individual’s health data along with assistance from the predictions based on modelled data. The cost of treatments can be reduced tremendously if no unnecessary tests are done. The project model first involves a tri-sensor system which takes the heartbeat, pulse-rate and oxygen saturation level of the patient. The values of all these three entities is displayed on the LCD Screen in the hardware model. The values of these sensor are able to be accessed from an Android App with the help of interfacing the App with the Hardware system using Arduino based coding. The LCD screen and Android App show the sensor data and the App data respectively which can be accessed by both the doctor and the patient’s family at any point of time simply by downloading the App. The complete history of the patient’s health is recorded in the IO server Adafruit. The MQTT protocol has been used for transferring information in sensor data from hardware to Adafruit and further publishing it from server to android app. MQTT uses a publish-subscribe methodology here with Adafruit as server while the hardware and Android App act as its two clients. The information is stored in Tabular form in Adafruit as well as in Graphical format. The graph form helps to see if any sudden discrepancies in values for sensor data occur and is used to warn.
Keywords: Adafruit, ESP Wi-Fi, Android App, MQTT, SVM