Identification of Respiratory Diseases and Diabetes by Non-invasive Method Using IoT

ABSTRACT

The objective of this paper is to identify respiratory diseases such as Asthma, Covid-19, pulmonary disease, and diabetes from the human breath odor using a non-invasive method. For detecting diseases using a non-invasive method, temperature sensor (to identify body temperature), pulse oximeter sensor (to identify blood oxygen level and heartbeat rate), and acetone sensor (to identify respiratory diseases from human breath odor) with Arduino ATMega328 microcontroller unit (MCU) were used. If the temperature is greater than 37.2 C, the heartbeat rate is greater than 100 bpm, and the oxygen level is less than 92% Covid-19 will be detected. If the oxygen level is less than 95% the heartbeat rate is at (100–125) bpm, and the temperature is at 36.1–37 C, asthma will be detected. If the heart rate is greater than 86 bpm, the temperature at 36.1–37 C, the oxygen level at 92–97% and the acetone level at (354–496) ppm, diabetes will be detected. If the oxygen level is less than 92% the temperature at 36.1–37 C, and the heartbeat rate is greater than 110 bpm, the pulmonary disease will be identified. After disease detection, suggestions will be provided to the patients based on their health reports. Finally, suggested medicines will be sent to the patient's registered mobile phones by connecting node MCU with blynk using IoT technology. The results will be stored and the patients can compare their health conditions for future analysis. The traditional method of laboratory tests is considered to consume more time. In our method, the duration of the detection process is less and the results help to identify health problems at early stages and predict diseases quickly compared to the traditional method. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.