Project Description
The aim of this project is to accomplish the completed design of a cardiovascular monitoring system, using data from two common sensors from the medical domain, namely an EKG (electrocardiogram), and also a stethoscope (implemented electronically). The monitoring system incorporates both EKG and stethoscope analysis, with the aim of determining the presence or absence of a set of medical conditions.
Operational Brief
There are two main areasto this project, the first being a wireless Embedded Transmission Unit (ETU)This combines the sensors, signal conditioning circuitry, a microcontroller and a wireless module.
Secondly, a remote PC server equipped with a matching wireless module receives the data from both sensors. The EKG data is analysed, comprising algorithmic feature extraction and a neural network classifier. The stethoscope audio is also analysed (for heart rate and murmur detection), by an algorithmic methodology. Both signals and analysis results are displayed graphically.
EKG classification is trained and evaluated on the MIT-BIH arrhythmia database, with correct classification of 93.72% of the test inputs.
No comments:
Post a Comment