Custom, spray coated receive coils for magnetic resonance imaging

www.nature.com/scientificreports OPEN Custom, spray coated receive coils for magnetic resonance imaging A. M. Zamarayeva1,2, K. Gopalan1,2*, J. R. Corea1, M. Z. Liu1, K. Pang1, M. Lustig1 & A. C. Arias1 We have developed a process for fabricating patient specific Magnetic Resonance Imaging (MRI) Radio-frequency (RF) receive coil arrays using additive manufacturing. Our process involves spray deposition of silver nanoparticle inks and dielectric materials onto 3D printed substrates to form high-quality resonant circuits. In this paper, we describe the material selection and characterization, process optimization, and design and testing of a prototype 4-channel neck array for carotid imaging. We show that sprayed polystyrene can form a low loss dielectric layer in a parallel plate capacitor. We also demonstrate that by using sprayed silver nanoparticle ink as conductive traces, our devices are still dominated by sample noise, rather than material losses. These results are critical for maintaining high Signal-to-Noise-Ratio (SNR) in clinical settings. Finally, our prototype patient specific coil array exhibits higher SNR (5 × in the periphery, 1.4 × in the center) than a commercially available array designed to fit the majority of subjects when tested on our custom neck phantom. 3D printed substrates ensure an optimum fit to complex body parts, improve diagnostic image quality, and enable reproducible placement on subjects. Healthcare technology can be significantly improved through customization to individual p atients1–3. Magnetic resonance imaging (MRI) is one of the examples where the customization of hardware could appreciably advance clinical outcomes. One of the key factors determining signal-to-noise ratio (SNR) of MR images is the design of the receive coils that are used to collect RF signal and their proximity to the patient6–8. Particularly, placing coils close to the body has been shown to significantly improve SNR and, thus, diagnostic image quality9,10. However, commercially available receive coils are typically designed to accommodate the largest possible subjects and do not optimally fit every patient or subject. This often results in substantial gaps between the coils and the body, which in turn compromises SNR. For instance, Corea et al. showed that placing coil only 1.8 cm away from the body results in an ~ 8% decrease in SNR9. Additionally, conventional coils are not designed for reproducible positioning on the patient, and do not restrict a patient from moving, which leads to motion artifacts during MRI scans. These shortcomings hinder the development of the next generation therapeutic approaches, such as MRI guided s urgeries11–13, that require multiple time-consuming MRI scans on a given patient visit. In addition, fMRI researchers who perform repeated scans on the same subject may find the reproducible placement of the coils beneficial. Custom receive coils that are fabricated on-demand to fit a patient’s or subject’s anatomy would address some of the aforementioned limitations. However, the established commercial manufacturing process is not suitable for on-demand and custom coil production. Typical coil manufacturing requires a trained RF engineer and involves hand assembly and packaging of electronic components such as copper wires and porcelain capacitors14. Entirely new approaches that allow seamless manufacturing and integration of electronic elements must be adopted to enable custom MRI coils. Novel additive manufacturing techniques and solution-processed materials offer a potential to transform patient-specific coil m anufacturing15,16. The common concern with shifting towards solution-processed electronic materials is the higher loss associated with their use; for example, printed solution-based conductors exhibit lower conductivities than that of bulk metals17. However, in clinical MRI intrinsic losses in the system are dominated by losses stemming from the human body18,19. Therefore, printed materials could perform comparably or better than the conventional materials, while enabling additive manufacturing of custom coils. The first demonstrations of the conformal MRI receive coils did not rely on additive manufacturing and were fabricated by sewing conductors into f abric20, using m ercury21,22, or copper t ape23–25 as a conductor. Mager et al. produced flexible coils using ink-jet printing26. While inkjet printing allows printing coils onto the flexible substrates, it requires many printing passes to achieve the desired conductivity for RF applications15. Corea et al. developed highly flexible and lightweight receive coils that were fabricated using scalable and low-cost 1 Department of Electrical and Computer Engineering, University of California Berkeley, Berkeley, CA, USA. 2These authors contributed equally: A. M. Zamarayeva and K. Gopalan. *email: Scientific Reports | (2021) 11:2635 | https://doi.org/10.1038/s41598-021-81833-0 1 Vol.:(0123456789) www.nature.com/scientificreports/ Figure 1.  Manufacturing flow diagram for custom specific MRI receive cells. (a) A scan of the volunteer’s body part (neck) generated using a structural scanner. (b) CAD drawing generated using the structural scan to perfectly fit the volunteer’s neck and used to 3D-print a custom substrate. (c) Schematic representation of spray deposing coil components onto the 3D-printed custom substrate. (d) Q spoiling and matching circuitry connected to the coil with plastic screws and conductive epoxy. screen-printing approach9,10,13. This paved the way to new opportunities in imaging, particularly for pediatric patients for whom conventional adult coils are especially problematic. In this work we developed a process for additive manufacturing of 3D patient-specific MRI coils. Such coils are advantageous for applications where, in addition to improved SNR and reproducible placement on the patient are important. The 3D coils also ensure perfect fit to the body parts with complex geometries, like a neck, which is challenging to achieve with flexible 2D coils. To demonstrate how custom 3D printed coils can improve clinical imaging, we manufactured a custom neck array for c-spine and carotid artery imaging. Commercial neck coils are positioned at a distance from the body (Fig. S3a) to have large field of view and fit the majority of subjects, at the expense of SNR. High-resolution neck imaging is an indispensable tool for the evaluation of health conditions involving the neck and cervical spine. For instance, lesions27–29 or plaque accumulation in the carotid artery leading to s troke28,30,31 could be imaged and detected. We used conventional and printed coil arrays to image a neck-shaped loading phantom and compared the SNR between the two coil arrays. The SNR measured with the printed array exceeded that of the commercially available four channel neck array (Siemens, Erlangen. Shown in Fig. 3f) by forty percent in the center of the phantom and up to five (...truncated)