How To Make Your Cell Tracker Say"I dunno!"

2025-03-12

AI Summary

In high-throughput live-cell microscopy, linear assignment-based tracking algorithms lack uncertainty quantification, compromising analytical reliability. To address this, we propose the first plug-and-play uncertainty calibration framework that unifies Bayesian inference and classification confidence estimation within the linear assignment paradigm. Our method is detector- and motion-model-agnostic, and natively compatible with modern trackers—including Transformer-based architectures. Extensive evaluation across multiple state-of-the-art tracking algorithms demonstrates well-calibrated uncertainty estimates (Expected Calibration Error < 0.05) and strong discriminative power (AUROC > 0.85). Consequently, it significantly improves robustness and interpretability in trajectory filtering, anomaly detection, and downstream quantitative analysis. This work establishes a reliable, uncertainty-aware tracking paradigm for high-throughput studies of cellular dynamics.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Motion & Tracking
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Security and Privacy: Tracking, profiling, and countermeasures against them
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Cell tracking is a key computational task in live-cell microscopy, but fully automated analysis of high-throughput imaging requires reliable and, thus, uncertainty-aware data analysis tools, as the amount of data recorded within a single experiment exceeds what humans are able to overlook. We here propose and benchmark various methods to reason about and quantify uncertainty in linear assignment-based cell tracking algorithms. Our methods take inspiration from statistics and machine learning, leveraging two perspectives on the cell tracking problem explored throughout this work: Considering it as a Bayesian inference problem and as a classification problem. Our methods admit a framework-like character in that they equip any frame-to-frame tracking method with uncertainty quantification. We demonstrate this by applying it to various existing tracking algorithms including the recently presented Transformer-based trackers. We demonstrate empirically that our methods yield useful and well-calibrated tracking uncertainties.

Problem

• Develop uncertainty-aware tools for automated cell tracking.
• Quantify uncertainty in linear assignment-based tracking algorithms.
• Apply Bayesian inference and classification to improve tracking accuracy.

Innovation

• Bayesian inference for uncertainty quantification
• Classification approach in cell tracking
• Framework for frame-to-frame tracking enhancement