Quantitative assessments of image intensifier distortion induced by weak (Sub‐Gauss) magnetic fields during fluoroscopically‐guided procedures

Abstract

Background: Fluoroscopically-guided procedures at our hospital have been aborted due to sigmoidal distortion (S-distortion) when an image intensifier (II) system is used in a surgical environment distant from any apparent sources of strong magnetic fields, such as a nearby magnetic resonance imaging (MRI) scanner. Clearly, current clinical practice fails to account for the impact of ambient weak magnetic fields and/or other contributing factors on S-distortion induction.Purpose: This study attempts to quantitatively assess the threshold level of magnetic field, along with other potential factors, that can induce intolerable S-distortion during image-intensified fluoroscopically-guided procedures. We will also discover the origins of such level of magnetic field in typical surgical facilities and provide our practical mitigation strategies accordingly. Methods: Ten surgical facilities and their accessory equipment (e.g., surgical tables) were screened using an AC/DC gaussmeter for the distribution and magnitude of magnetic field (magnetic flux density). A 'hot spot' of magnetic field was identified to further investigate the induction of S-distortion by scanning a titanium rod phantom using a GE OEC 9900 Elite II system placed at increasing distance from the 'hot spot' corresponding to decreasing magnetic field experienced by the II. The measurements were compared to that on a 'cold spot', and a GE flat panel detector (FPD) fluoroscopy was used as the negative control. Rod phantoms made of various magnetic susceptible materials (titanium, steel, aluminium, and copper) were scanned to explore the potential effects of implant material on S-distortion. An upper extremity anthropomorphic phantom was imaged on various surgical tables to mimic clinical sceneries. The GE II model and Siemens ARCADIS Orbic II model were compared to evaluate if S-distortion induction varied among different II models. Two metrics, angle of rotation (θ) and deviation/length ratio, were used to quantify the degree of S-distortion. Three designs of external magnetic shielding were evaluated for mitigating S-distortion. Results: We identified static magnetic fields up to 2500 µT and 70 µT on the floor and at 1-meter height, respectively, in random locations of surgical facilities. A large variation of magnetic field (64 ± 20 µT) was detected on the surface of surgical tables, with background magnetic fields of ∼35 µT. Quantitative assessments demonstrated that even weak magnetic fields at sub-Gauss level (<100 µT) could induce noticeable distortion artifacts, deemed unacceptable (θ > 4 • ).S-distortion was independent of the implant material being imaged but dependent on the II model -the threshold magnetic fields (4 • distortion induction) were as low as 47 µT and 94 µT for the GE and Siemens II models.