Abstract:
Over the past several years, the adoption of Internet of Things (IoT) technology
in healthcare has facilitated the continuous tracking of patients' vital signs. As a result,
the early detection of health emergencies is possible. This system introduces a real-time
IoT-based system that senses the vital signs of the body and carries out the diagnosis
through the MAX30102 sensor and Arduino Nano ESP32 microcontroller. The system
obtains the primary vital signs of the human body like Heart Rate (HR) and blood
oxygen saturation (SpO₂) and after that, it transmits the values via wireless
communication to the ThingSpeak cloud platform for the graphical representation and
remote access. Employing a rule-based classification algorithm, the analyzed values are
divided into different health classes that the computer recognizes based on the input
like Normal, Low, and Abnormal, thus allowing timely alerts for possible health risks.
The system is also characterized by the use of classification metrics such as
accuracy, precision, recall, F1-score, and specificity, along with real-time performance
indicators including latency and data transmission reliability to evaluate its
performance. The experiments’ results corroborate that the designed system reaches
high accuracy, quickness, and stability. In other words, the device is excellent for
continuous patient monitoring and early emergency detection. This method provides a
less costly, smaller-sized, and less energy-consuming system for health applications
based on IoT, which may lead to the improvement of the patient's health and taken steps
in time to prevent disease progression. Experimental results show that the proposed
system achieves an accuracy of 90%, precision of 89%, recall of 67%, a Mean Absolute
Error (MAE) of 1.35, and a Root Mean Square Error (RMSE) of 1.75 in detecting and
estimating heart rate values compared with references for timestamped storage.