Marker-Less Lung Tumor Tracking from Real-Time Color X-Ray Fluoroscopic Images Using Cross-Patient Deep Learning Model

TL;DR

This paper introduces a non-invasive method for tracking lung tumors during radiotherapy using deep learning, eliminating the need for implanted markers.

Contribution

A cross-patient deep learning model is proposed for marker-less tumor tracking, avoiding per-patient retraining.

Findings

The framework achieved a median 3D tumor center tracking error of 1.53 mm.

Average processing time was 179.8 ms per image, showing real-time feasibility.

Directional errors were within clinically acceptable ranges for all axes.

Abstract

Fiducial marker implantation for tumor localization in radiotherapy is effective but invasive and carries complication risks. To address this, we propose a marker-less tumor tracking framework to explore the feasibility of a cross-patient deep learning model, aiming to eliminate the need for per-patient retraining. A novel degradation model generates realistic simulated data from digitally reconstructed radiographs (DRRs) to train a Restormer network, which transforms clinical fluoroscopic images into clean, DRR-like images. Subsequently, a DUCK-Net model, trained on DRRs, performs tumor segmentation. We conducted a feasibility study using a clinical dataset from 7 lung cancer patients, comprising 100 distinct treatment fields. The framework achieved an average processing time of 179.8 ms per image and demonstrated high accuracy: the median 3D Euclidean tumor center tracking error was 1.53 mm, with directional errors of 0.98±0.70 mm (LR), 1.09±0.74 mm (SI), and 1.34±0.94 mm (AP). These promising results validate our approach as a proof-of-concept for a cross-patient marker-less tumor tracking solution, though further large-scale validation is required to confirm broad clinical applicability.

Figures