Method and System for Tracking of a Virtual Electrode on a Coronary Sinus Catheter in Fluoroscopic Images

An apparatus for tracking a virtual electrode (VE) on a coronary sinus (CS) catheter in a fluoroscopic image sequence comprising:

• means for detecting a VE part of the CS catheter in the first frame based on input locations of a plurality of CS catheter electrodes and an input location of VE in the first frame, wherein the VE part is a segment of the CS catheter between a most proximal electrode and the VE;
• means for initializing a catheter electrode template and a VE part template in a first frame of a fluoroscopic image sequence based on the input locations of the plurality of CS catheter electrodes and the input location of the VE in the first frame;
• and
• means for tracking the VE in a second frame of the fluoroscopic image sequence comprising:
• means for detecting electrode position candidates and catheter body point candidates in the second frame of the fluoroscopic image sequence using respective trained detectors;
• means for tracking the catheter electrode template in the second frame based on the detected electrode position candidates;
• means for generating VE part hypotheses based on the most proximal electrode in the catheter electrode template in the second frame;
• means for calculating a probability score for each of the VE part hypotheses;
• and
• means for selecting one of the VE part hypotheses as the VE part result in the second frame based on the probability score.

A non-transitory computer readable medium encoded with computer executable instructions for tracking a virtual electrode (VE) on a coronary sinus (CS) catheter in a fluoroscopic image sequence, the computer executable instructions defining a method comprising:

• receiving user inputs indicating locations of a plurality of CS catheter electrode and a location of a VE in a first frame of the fluoroscopic image sequence;
• detecting a VE part of the CS catheter in the first frame, wherein the VE part is a segment of the CS catheter between a most proximal electrode and the VE;
• initializing a catheter electrode template and a VE part template in a first frame of a fluoroscopic image sequence based on the input locations of the plurality of CS catheter electrodes and the input location of the VE in the first frame;
• and
• tracking the VE in a second frame of the fluoroscopic image sequence by:
• detecting electrode position candidates and catheter body point candidates in the second frame of the fluoroscopic image sequence using respective trained detectors;
• tracking the catheter electrode template in the second frame based on the detected electrode position candidates;
• generating VE part hypotheses based on the most proximal electrode in the catheter electrode template in the second frame;
• calculating a probability score for each of the VE part hypotheses;
• and
• selecting one of the VE part hypotheses as the VE part result in the second frame based on the probability score.