Intelligent Risk Alarm for Asthma Patients using Artificial Neural Networks

TL;DR

This paper presents an intelligent alarm system for asthma patients that uses artificial neural networks to analyze atmospheric chemicals and environmental factors, providing early warnings based on sensor data with high accuracy.

Contribution

It introduces a novel sensor-based wristwatch system utilizing ANN to predict asthma risk from atmospheric chemicals and environmental conditions, achieving 99.58% accuracy.

Findings

Achieved 99.58% classification accuracy.

Demonstrated effective detection of harmful atmospheric chemicals.

Proposed a wearable system for real-time asthma risk warning.

Abstract

Asthma is a chronic disease of the airways of the lungs. It results in inflammation and narrowing of the respiratory passages, which prevents air flow into the airways and leads to frequent bouts of shortness of breath with wheezing accompanied by coughing and phlegm after exposure to inhalation of substances that provoke allergic reactions or irritation of the respiratory system. Data mining in healthcare system is very important in diagnosing and understanding data, so data mining aims to solve basic problems in diagnosing diseases due to the complexity of diagnosing asthma. Predicting chemicals in the atmosphere is very important and one of the most difficult problems since the last century. In this paper, the impact of chemicals on asthma patient will be presented and discussed. Sensor system called MQ5 will be used to examine the smoke and nitrogen content in the atmosphere. MQ5 will be inserted in a wristwatch that checks the smoke and nitrogen content in the patients place, the system shall issue a warning alarm if this gas affects the person with asthma. It will be based on the Artificial Neural Networks (ANN) algorithm that has been built using data that containing a set of chemicals such as carbon monoxide, NMHC (GT) acid gas, C6H6 (GT) Gasoline, NOx (GT) Nitrogen Oxide, and NO2 (GT) Nitrogen Dioxide. The temperature and humidity will be also used as they can negatively affect asthma patient. Finally, the rating model was evaluated and achieved 99.58% classification accuracy.