Advancing intraoperative magnetic tracing using 3D freehand magnetic particle imaging

Abstract

Purpose: Sentinel lymph node biopsy is a routine procedure for nodal staging in penile cancer. Most commonly, this procedure is guided by radioactive tracers, providing various forms of preoperative and intraoperative guidance. This is further extended with fluorescence imaging using hybrid radioactive-fluorescence tracers. Alternatively, a magnetic-based approach has become available using superparamagnetic iron-oxide nanoparticles (SPIONs). This study investigates a novel freehand magnetic particle imaging and navigation modality (fhMPI) for intraoperative localization, along with a hybrid approach, combining magnetic and fluorescence guidance.

Materials and methods: The fhMPI set-up was built with a surgical navigation device, optical tracking system and magnetometer probe. A dedicated reconstruction software based on a look-up-table method was used to reconstruct a superficial 3D volume of the SPION distribution in tissue. For fluorescence guidance, indocyanine green (ICG) was added to the SPIONs. The fhMPI modality was characterized in phantoms, ex vivo human skin and in vivo porcine surgery.

Results: Phantom and human skin explants illustrated that the current fhMPI modality had a sensitivity of 2.2 × 10^-2 mg/mL SPIONs, a resolving power of at least 7 mm and a depth penetration up to 1.5 cm. Evaluation during porcine surgery showed that fhMPI allowed for an augmented reality image overlay of the tracer distribution in tissue, as well as 3D virtual navigation. Besides, using the hybrid approach, fluorescence imaging provided a visual confirmation of localized nodes.

Conclusion: fhMPI is feasible in vivo, providing 3D imaging and navigation for magnetic nanoparticles in the operating room, expanding the guidance possibilities during magnetic sentinel lymph node procedures. Furthermore, the integration of ICG provides the ability to visually refine and confirm correct localization. Further clinical evaluation should verify these findings in human patients as well.

Keywords: Augmented reality; Fluorescence-guided surgery; Magnetic particle-guided surgery; Penile cancer; Surgical navigation.

Fig. 1

Schematic overview of the fhMPI imaging and navigation set-up. a Overview of the complete set-up, displaying the surgical navigation device, the handheld magnetometer probe and the near-infrared (NIR) optical tracking. b Magnification of the handheld magnetometer probe, displaying the probe reference target (RT^probe) and the patient reference target (RT^patient)

Fig. 2

Visualization of the size-exclusion column measurements for ICG-SPION. Fluorescence intensity is shown of the first fraction after purification of a column with ICG and a column with ICG-SPION. This clearly shows a rise of fluorescence emission at 810 nm for ICG-SPION, indicating that at least part of the ICG has bound non-covalently to the SPION particles. The intensity values were corrected by subtracting the SPION background intensity and normalized to that of ICG-SPION

Fig. 3

Ex vivo human skin evaluation of the fhMPI modality. a Augmented reality overlay of the fhMPI scan on the target anatomy. The calculated distance between the magnetometer probe tip and the centre of the tracer hotspot that is pointed at, is displayed in the upper right corner of the image. b Virtual reality navigation towards the tracer hotspots detected in the image, again with the distance towards the hotspots displayed in the upper right corner of the image. c Confirmation of the surgical target using fluorescence imaging

Fig. 4

Characterization of the fhMPI modality. a fhMPI sensitivity, displaying the minimal SPION concentration needed for successful imaging. b fhMPI depth penetration, showing successful scanning is still possible up to 1.5 cm. c Overview of the resolving power results, showing that two lymph-node-like targets were still resolvable when the edges of the nodes were at a distance of at least 7 mm

Fig. 5

In vivo porcine evaluation of the fhMPI modality. a Targeted SLNs displayed within the anatomy. b Surgical localization starts with rough magnetic tracing. c This is followed with a fhMPI scan in the area of interest, displaying the SLN location registered as an augmented reality overlay on the anatomy. d Virtual reality navigation towards the lymph node location. e Fluorescence imaging confirms the lymph node location once the tissue is exposed. f Ex vivo evaluation depicts a typical SPION-brownish colour in the specimen. g SPION uptake confirmed with fhMPI. h SPION uptake confirmed with fluorescence. i Histopathology confirms the SPIONs (in blue) are indeed in the resected lymph node