354 - Intrafractional Markerless Lung Tumor Tracking: First Clinical Experience with AI-Empowered Target Decomposition Technique
Presenter(s)
Y. Fu1, P. Zhang1, Q. Fan1, W. Cai1, H. Pham1, S. Burleson1, N. Shaverdian2, A. J. Wu2, L. I. Cervino1, J. M. Moran1, T. Li1, and X. Li1; 1Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, 2Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY
Purpose/Objective(s): This study presents the first clinical experience of intrafractional markerless lung tumor tracking enabled by an AI-driven target decomposition technique. The primary objectives are to characterize intrafractional lung tumor motion and assess the feasibility of reducing PTV margins in deep inspiration breath hold (DIBH) lung SBRT patients.
Materials/Methods: Fifteen lung SBRT patients were enrolled and treated on a medical linear accelerator platform under DIBH conditions, receiving 3–5 fractions with respiratory motion managed using a Varian RPM system with a 3 mm gating window. During treatment planning, a patient-specific deep learning model was trained on simulation CT scans to enhance tumor contrast in kV projection images via a target decomposition technique. The model was integrated into our clinical software that utilizes template matching to track tumor motion on intrafraction motion review (IMR) images triggered every 200 monitor units during beam delivery. A qualified medical physicist manually verified tracking quality, with inaccurate data excluded from the analysis. Tumor motion was quantified by calculating the mean and standard deviation of the maximum displacement from the isocenter in both the longitudinal and in-plane left-right (IPLR) directions. The percentage of treatment time during which tumor displacement remained below thresholds of 5, 4, 3, and 2 mm were calculated.
Results: Tumor tracking was successfully performed on 1,222 IMRs out of 1269 IMRs collected across 56 treatment sessions, achieving a 96.3% tracking rate. Tracking on 47 IMRs was excluded from analysis due to either inaccurate target decomposition or inaccurate tumor template matching. The average maximum tumor displacement from the isocenter was 3.3±2.1 mm in the longitudinal direction and 2.8±1.7 mm in the IPLR direction. On average, tumor displacement remained below 5, 4, 3, and 2 mm for 94.9%, 91.5%, 84.5%, and 69.9% of the treatment time, respectively.
Conclusion: This study represents the first clinical implementation of intrafractional markerless lung tumor tracking using an AI-driven target decomposition technique. Our findings suggest that, when combined with appropriate tumor motion compensation strategies (e.g., couch compensation), this method could facilitate PTV margin reduction for DIBH lung SBRT patients.