Robust Tracking with Particle Filtering for Fluorescent Cardiac Imaging

TL;DR

This paper introduces a particle filtering tracking method with cyclic-consistency checks for fluorescent cardiac imaging, enabling real-time, robust tracking of multiple targets despite heart motion and image fluctuations.

Contribution

The proposed cyclic-consistency particle filtering approach improves robustness and accuracy in tracking cardiac features during surgery, outperforming existing deep learning and conventional methods.

Findings

Tracks 117 targets at 25.4 fps in real-time

Achieves a tracking error of 5.00 pixels

Outperforms other trackers in accuracy

Abstract

Intraoperative fluorescent cardiac imaging enables quality control following coronary bypass grafting surgery. We can estimate local quantitative indicators, such as cardiac perfusion, by tracking local feature points. However, heart motion and significant fluctuations in image characteristics caused by vessel structural enrichment limit traditional tracking methods. We propose a particle filtering tracker based on cyclicconsistency checks to robustly track particles sampled to follow target landmarks. Our method tracks 117 targets simultaneously at 25.4 fps, allowing real-time estimates during interventions. It achieves a tracking error of (5.00 +/- 0.22 px) and outperforms other deep learning trackers (22.3 +/- 1.1 px) and conventional trackers (58.1 +/- 27.1 px).