Reaction to Science Advances Editorial Detection of stroke by portable, low-field MRI: A milestone in medical imaging
Author: Peter Basser, Section on Quantitative Imaging and Tissue Sciences, NICHD, NIH, Bethesda, MD, USA
Science Advances, 20 Apr 2022, Vol 8, Issue 16, https://doi.org/10.1126/sciadv.abp9307
Portable, low-field magnetic resonance imagers can aid the clinical assessment of stroke. They may also help democratize access to scarce medical imaging resources.
Q&A with Hyperfine Thought Leaders
We are here with Hyperfine Chief Product Officer Tom Teisseyre and Hyperfine Senior Medical Director Chip Truwit, MD, to get their thoughts on the recent editorial “Detection of stroke by portable, low-field MRI: A milestone in medical imaging” from Peter Basser, NICHD in the April 20 issue of Science Advances.
Hyperfine: Tom, as the chief product officer of the only commercially available portable, low-field MR imaging system, what encourages you the most about the medical community’s excitement around portable low-field magnetic resonance imaging?
Tom Teisseyre: This is quite personal. When I finished grad school, the job offers and opportunities were from big companies making small changes. MRI has been going through a period of incremental improvements, enlarging systems, ballooning costs, and a focus on resolution. The learning at Hyperfine is that contrast and expanding the clinical utility of portable MRI to potentially include use cases for ischemic stroke are equally important. Advances in computing, image processing, and AI can allow a portable MRI system to provide images that can answer key clinical questions with significantly less capital overhead. This is a case of “there’s plenty of room at the bottom” (Richard Feynman). In the editorial discussed here, Basser jumps right into the cool waters of why this space is ripe for disruption right from the intro.
- “From a public health perspective, although 3- and 7-T scanners can produce exquisite brain images, this capability comes at a cost.”
- “For some indications, including ischemic stroke, these MRI scanners are a welcomed addition to the clinical armamentarium, as they have the potential to improve some aspects of clinical care over the current standard of care.”
- “Owing to their reduced cost and portability, these scanners could be deployed in a myriad of new settings.”
- “Future innovations in motion correction, noise remediation, and image data upload capabilities suggest the eventual use of these scanners in ambulances or even on the battlefield.”
Teisseyre: The only disagreement I have with Basser is that we at Hyperfine are already using advanced hardware and software today to increase the portability and reduce the costs of MRI. There are many untapped and future innovations in motion correction and noise cancellation, as Basser states, but the fundamental transformational technology already exists in the Hyperfine Swoop® Portable MR Imaging System™. Without it, imaging at low field and the images in the Sheth paper would not be possible. Future innovation is happening today. Notably, the images in Sheth’s paper are from four subsequent generations of software (when we first built the enabling software and algorithms) and well before our deep learning pipeline. Today you can see the impact of the work with significantly better images, and it’s just the beginning.
Basser editorial: “Low-field portable MRI of stroke is not without competition in the low-cost, rapid diagnostic imaging arena.”
Teisseyre: I couldn’t agree more about CT being a key part of the armamentarium. Even today, with CTs being ubiquitous, some 80% of stroke patients go untreated in the US, including in top medical centers1,2. This is why we focused our development on DWI. CT Perfusion is a surrogate secondary metric for what you see with MRI. With DWI and FLAIR contrast, clinicians can see directly viable tissue and make an action. We believe that by making DWI ubiquitous and collaborating with our clinical partners, we’ll be able to make a dent in that 80%.
Basser editorial: “One way to hasten these needed developments is for portable, low-field MRI manufacturers to provide image acquisition and processing software, hardware specifications, and pulse sequences to the user community, i.e., MRI scientists, inventors, students, radiologists, emergency room physicians, and engineers.”
Teisseire: We’re here to support the MRI community. The Swoop system provides an “expert environment” for MRI scientists, inventors, students, and engineers to help us drive this revolution together. This is just the beginning. Let’s get started.
Hyperfine: Chip, as a neuroradiologist who has transitioned into the industry in just the past several years, first for a conventional MRI manufacturer and now for Hyperfine, what points mentioned in this editorial do you hear echoed the most in conversations with clinicians both in the US and around the world?
Chip Truwit: There were three sections from the editorial that resonated with what I hear from clinicians.
Basser editorial: “If urgent care facilities or local hospital emergency rooms offered portable low-field brain MRI as readily as they offer ultrasound imaging, for instance, then it would help bridge this gap in patient access to emergency medical care.”
Truwit: Access to MR while in the ED continues to be of great importance to patients and providers alike. Siting issues and the more vexing problem of finding space continue to pose barriers to everyone’s best intentions. Swoop addresses both the space and siting issues.
Basser editorial: “Low-field head scanners could also serve a range of patients who could not be scanned with conventional or high-field MRI scanners.”
Truwit: Radiologists and clinicians have been frustrated that the medical device industry has been slow to solve MR safety and compatibility concerns around their devices. The Swoop system potentially addresses those issues. But as the first portable MRI, medical device manufacturers have not yet published individual device testing for the low field MRI environment.
Basser editorial: “Portable, low-field scanners, with open designs, would expand pediatric brain imaging research opportunities with an expected higher reward-to-risk ratio than at higher field strengths.”
Truwit: While the signal is obviously less than at 1.5T, let alone at 3T or 7T, what has become apparent is that signal is only one of many components that determine whether an image can be obtained and how good the image can be. Who would have predicted that this would work? Who would have predicted that noise cancellation would be so valuable? Who would have predicted that AI would be so impactful?
1. Fadar Oliver Otite, Vasu Saini, Nicole Beaton Sur, Smit Patel, Richa Sharma, Emmanuel O. Akano, Nnabuchi Anikpezie, Karen Albright, Elena Schmidt, Haydn Hoffman, Grahame Gould, Priyank Khandelwal, Julius Gene Latorre, Amer M. Malik, Ralph L. Sacco and Seemant Chaturvedi (2021). Ten-Year Trend in Age, Sex, and Racial Disparity in tPA (Alteplase) and Thrombectomy Use Following Stroke in the United States, Stroke, 52:2562-2570. Retrieved on May 11, 2022, from https://www.ahajournals.org/doi/10.1161/STROKEAHA.120.032132
2. Anand, S. K., Benjamin, W. J., IV, Adapa, A. R., Park, J. V., Wilkinson, D. A., Daou, B. J., Burke, J. F., & Pandey, A. S. (2021). Trends in acute ischemic stroke treatments and mortality in the United States from 2012 to 2018, Neurosurgical Focus, 51(1), E2. Retrieved May 11, 2022, from https://thejns.org/focus/view/journals/neurosurg-focus/51/1/article-pE2.xml