Healthcare Professionals / Electrophysiology / LUX-Dx™ ICM / Reject False Positives
Our Dual-Stage ICM Algorithm Tests Itself, Not You
The LUX-Dx™ Insertable Cardiac Monitor (ICM) System uses a dual-stage algorithm that automatically detects and then verifies data before sending results, so it can reject false positives, provide actionable insights and get you to an accurate diagnosis faster.
It’s time for data that works. It’s time for something different.
An ICM That Checks its Work
reject false positives and see performance results
across multiple arrythmias for LUX-Dx ICM.
LUX-Dx ICM Algorithms Advanced Algorithms for Enhanced Accuracy |
AF Algorithm
Monitors R-R variability to detect potential AF rhythms. The verification stage uses adaptive morphology, noise discrimination and pattern detectors to identify and reject false positives.
AT Algorithm
Uses rate, duration and built-in flexibility to detect only high-rate rhythms sustained for a prolonged period of time, or to detect short duration AT/Atrial Flutter. The AT algorithm is independent of AF algorithm.
Pause Algorithm
Monitors R-R duration to detect pause episodes. The verification stage rejects false positives by using a dynamic noise-reduction filter, signal-to-noise ratio and loss-of-signal conditions.
Brady Algorithm
Uses rate and duration parameters during the detection stage to identify potential brady episodes. During the verification stage, episodes are further examined for under-sensing before being rejected.
Tachy Algorithm
Uses traditional ICD-based rate and duration parameters during the detection stage. The verification stage uses a machine-learning-based decision tree to identify potential tachy episodes. If they are not within the rate zone, they’re rejected as noise.
Bench Testing
Algorithm Performance
Download the HRS posters below to see how the LUX-Dx ICM algorithm performed in bench testing.*
Improved AF Rhythm Discrimination with an Implantable Cardiac Monitor Using QRS Morphology
See how using QRS morphology assessment in addition to R-R variability resulted in a 53.1% relative reduction in false positives and AF burden Positive Predictive Value (PPV) ranged from 73% to 99%. 1
A Novel Algorithm to Improve Atrial Fibrillation Detection in Implantable Cardiac Monitors
Explore the research showing PPV improved from 54.4% to 99.2% with minimal changes to sensitivity. 2
A Novel Algorithm Reduces False Positives for Pause Detection in Implantable Cardiac Monitors
See the data showing sensitivity performance for three-second pauses from four different settings ranged from 92% to 99% and PPV ranged 82% to 98%. 3
A Novel Algorithm Improves Detection of Arrhythmias with Regular RR Intervals in Implantable Cardiac Monitors
See the data demonstrating PPV for combined AF and AT detection increased from 54% to 97% with both higher rates and longer durations. 4
Can Machine Learning Be Used to Optimize a Tachycardia Detection Algorithm in an Implantable Cardiac Monitor?
Review the data showing how machine learning correctly rejected noise 96% of the time, which improved the PPV from 23% to 86%. 5
Bench Testing One-Pager
Download a summary of LUX-Dx ICM bench testing results across multiple arrhythmias, including AF, AT, tachy and pause.
Stay Up to Date
Explore videos about the LUX-Dx ICM procedure, algorithms and more.
Product Details
Explore LUX-Dx ICM product specs, implant information and more.
References:
1. Mittal S, Saha S, Perschbacher D, Siejko K. Improved AF Rhythm Discrimination with an Implantable Cardiac Monitor Using QRS Morphology. Poster presented at: 2019 Heart Rhythm Society; May, 2019; San Francisco, CA.
2. Richards M, Perschbacher D, Herrmann K, Siejko K, Saha S. A Novel Algorithm Reduces False Positives for Pause Detection in Implantable Cardiac Monitors. Poster presented at: 2019 Heart Rhythm Society; May, 2019; San Francisco, CA.
3. Richards, M. Perschbacher, D, Herrmann, K, Saha, S, Siejko, K. A Novel Algorithm to Reduce False Positives for Pause Detection in Implantable Cardiac Monitors. Poster presented at Heart Rhythm Society May 2018, Boston, MA.
4. Richards, M. Perschbacher, D, Saha, S. A Novel Algorithm Improves Detection of Arrythmias With Regular R-R Intervals. Poster Presented at Heart Rhythm Society May 2019 San Francisco, CA.
5. Mittal, S. Siejko, K, Saha, S, Herrmann, K, Perschbacher, D. Can Machine Learning Be Used to Optimize Tachycardia Detection Algorithm in an Implantable Cardiac Monitor. Poster presented at ESC, June 2018, Munich, Germany.
*Bench Test results may not necessarily be indicative of clinical performance. Bench data for this research was provided by Telemetric and Holter ECG Warehouse (THEW), University of Rochester, NY.