Kinetic™ ECG Algorithm

Kinetic™

ECG Algorithm

The Kinetic™ ECG Algorithm provides signal processing and interpretation of the ECG waveform, thereby aiding health care professionals in assessing cardiac parameters. It provides highly accurate QRS detection and feature extraction, beat classification, interval measurement, and rhythm interpretation for up to sixteen leads of captured ECG data.

Kinetic™ is so versatile, it enables real-time or historical analysis in platforms ranging from in-hospital or call center PC or Server-based devices to small, patient-worn ambulatory devices that employ microcontrollers or DSP’s. The various Kinetic algorithms are easily combined and integrated together to provide you the functionality you need. Some key features of Kinetic™ are:

• Highly accurate

• Efficient code

• minimal battery usage

• Real-time

• No warm-up period

• On-line beat-by-beat output

• Analyze up to 16 leads

• T wave morphology changes

• Beat classification

• Arrhythmia detection

• Annotation file for further offline analysis

• Smoothed output for GUI usage

Kinetic™ is based on a novel method for detection of the QRS complex. At the core of the Kinetic™ ECG Algorithm is a unique waveform filter. This filter identifies the QRS by its intrinsic properties, which remain the same regardless of the wave’s changing morphology and rhythm patterns. Kinetic is able to process ECGs in real time, with no warm-up period, template matching routines, or manual intervention. Just automatic, precise analysis – from the first beat. 

The application and processor used will determine the optimal configuration of the Kinetic™ ECG Algorithm. Monebo will work with you to determine the best combination of features and system design based on your individual needs. An early version of the Algorithm received 510(k) clearance in 2007. The current version, which incorporates many enhanced features must be cleared in your device. The FDA considers the algorithm to be a “component”.

Algorithm performance is affected by equipment design, system noise, transmission method, environment, different rules or definitions of measurements and arrhythmias, and many other factors. Monebo has identified 54 configuration variables that can affect performance and has developed an Algorithm Optimization System to determine the optimal set of 54 parameter for specific acquisition devices and targeted arrhythmias. The system utilizes a genetic algorithm, a form of machine learning, to optimize the variables automatically, providing a comprehensive and thorough optimization, resulting in the best overall performance in your particular design.

Automatic, precise analysis - from the first beat