Cardiac-gated dual-energy imaging of the chest: Design and performance evaluation of a cardiac trigger based on a fingertip pulse oximeter

Purpose: A research prototype for high‐performance dual‐energy (DE) imaging of the chest is under development. This paper discusses the development and characterization of a cardiac gating system designed to precisely trigger the imaging system according to cardiac phase and minimize anatomical misregistration due to heart motion. Method and Materials: A fingertip pulse oximeter was employed to measure the peripheral pulse waveform and trigger x‐ray exposures during the quiescent phase of the heart (diastole). Temporal delays accounted in the timing scheme include physiological pulse propagation, waveform processing, and imaging system delays (filter‐wheel, bucky‐grid, and flat‐panel detector). An empirical model of the diastolic period allows calculation of the implemented delay, timp, required to trigger correctly at any patient heart‐rate. Performance was evaluated in terms of accuracy and precision of diastole‐trigger coincidence and expert assessment of cardiac motion artifact in gated and ungated DE images. Results: The model suggests a triggering scheme characterized by two heart‐rate (HR) regimes: below a HR‐threshold, sufficient time exists to expose on the same heartbeat [formula omitted]; above the HR‐threshold, a characteristic timp(HR) delays exposure to the subsequent heartbeat, accounting for all fixed and variable system delays. Initial implementation indicated 83% accuracy in diastole‐trigger coincidence. By modifying the HR estimation method (reduced temporal smoothing of the pulse waveform), trigger accuracy of 100% was achieved. Cardiac‐gated DE patient images demonstrate significantly reduced cardiac motion as assessed by expert radiologists. Conclusion: A pulse oximeter combined with a cardiac model provides accurate x‐ray triggering and significantly reduces heart motion artifacts. A simple fingertip clip presents logistic, cost, and workflow advantages compared to ECG. The system has been implemented in a clinical research trial, with gated and ungated arms allowing characterization of the impact of cardiac motion artifact on diagnostic performance. Conflict of Interest: Research sponsored in part by Eastman Kodak. © 2007, American Association of Physicists in Medicine. All rights reserved.