Browse Prior Art Database

A Tuning Method for Arrhythmia Detection using Camera Devices

IP.com Disclosure Number: IPCOM000239676D
Publication Date: 2014-Nov-24
Document File: 6 page(s) / 681K

Publishing Venue

The IP.com Prior Art Database

Abstract

Remote sensing image analysis is increasingly used to characterize the physical condition of medical patients. This idea proposes a method used to detect and characterize a patient’s atrial fibrillation using a video camera. The proposal is a method to automatically tune the parameters of an arrhythmia detection algorithm when applied to a video plethysmography (VPG) signal obtained from a video camera. The parameter estimation can be performed automatically in the hospital on patient data before the patient is sent home. The approach uses a performance function that quantifies the error between the VPG signal and any of the Electrocardiography (ECG) or photoplethysmography (PPG) signals. In the tuning process, ECG or PPG signals are used as reference signals along with the VPG signal. All signals are synchronized before performing tuning.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 26% of the total text.

A Tuning Method for Arrhythmia Detection using Camera Devices

Remote sensing image analysis is increasingly used to characterize the physical condition of medical patients.  This idea proposes a method used to detect and characterize a patient’s atrial fibrillation using a video camera.  The proposal is a method to automatically tune the parameters of an arrhythmia detection algorithm when applied to a video plethysmography (VPG) signal obtained from a video camera. The parameter estimation can be performed automatically in the hospital on patient data before the patient is sent home. The approach uses a performance function that quantifies the error between the VPG signal and any of the Electrocardiography (ECG) or photoplethysmography (PPG) signals. In the tuning process, ECG or PPG signals are used as reference signals along with the VPG signal. All signals are synchronized before performing tuning.

1. Summary

In this idea we describe a method for automatically tuning the parameters of an arrhythmia detection algorithm using reference signals from ECG data and a performance function. A segment of manual/auto annotated ECG data (15sec – 30sec) or a PPG sensor serves as the ground truth of peak-to-peak intervals. From each of the ECG/PPG interval’s, we identify the desired segment for tuning. Then, we apply the automatic peak detection algorithm with initial parameters to estimate all the peak-to-peak intervals within the segment. We use the sum of the squares of the error between standard deviations of pulse peak points and ECG RR intervals (or PPG pulse points) as the performance function. Calculate the performance function and iterate the process over and over with a different parameter set each time until the function is minimized. This method can be used in hospitals to customize the parameters for each patient before they are sent home.

2. Introduction and Prior Art

Atrial fibrillation (A-Fib) can cause palpitations, fainting, chest pain, congestive heart failure, and stroke. It is one of the most common sustained arrhythmias in adults. Its occurrence increases with age and presents with a wide spectrum of symptoms and severity. It is estimated that more than 2.4 million people in the United States are affected by A-Fib and routinely visit the hospital.

In A-Fib, the electrocardiographic signal (ECG) is characterized by irregular R-R intervals due to irregular conduction of impulses to the ventricles. A-Fib may occur in episodes lasting from minutes to days intermittently ("paroxysmal"), all the time (“persistent” or “chronic”) or be permanent (cannot be converted to normal) in nature.

The presence of A-Fib is detected using a body surface worn ECG. The absence of P-waves together with irregular intervals between ventricular contractions is in general a good marker of the presence of this arrhythmia. However, the use of the ECG to continuously monitor patients suspected of A-Fib in ambulatory condition remains a challen...