Early Chronic Pulmonary Disease Prediction System Using Machine Learning

In this project motive is how to predict early stage in COPD disease. It know about chronic is a longterm disease. It can be used to respiration sensor, gas sensor, smoke sensor, transformer, ArduinoUno, IoT module and LCD display. Smart breath analyzers are becoming into sophisticateddiagnostic instruments. Respiration sensor is difficult to breath to be utilized. The gas sensor isdetected to the gases in the environment. Arduino Uno is a controller it can be control with all thesensors. IoT module is collected all the sensor values. The LCD display shows to the sensor values.These gadgets allow for prompt medical intervention by continuously monitoring health conditionsand forecasting when illness like Chronic Pulmonary Disease (COPD) will worsen. The systemanalyses patterns in the concentrations of gas, smoke and acetone from exhaled breath that iscollected using a sampling bag and assessed using a gas-mixing bench using machine learningalgorithms based on decision trees. The incorporation of Decision Tree algorithms facilities realtime pattern recognition and anomly detection, providing personalized health insights for high-riskpatients. The models were trained on environmental and patient-specific datasets to enhanceprediction accuracy. Prototypes incorporating a arduino uno were tested in diverse settings,including indoor and outdoor environments. The findings show that ML-driven smart breathanalyzers enhance proactive healthcare and contribute to a more dependable and efficienttelemedicine ecosystem by improving early COPD detection and monitoring.