ReviewfMRI at 20: Has it changed the world?
Introduction
Functional MRI brings physics and technology face-to-face with psychology, neuroscience, and medicine. This juxtaposition, whereby each domain has driven developments in the other, has been a source of excitement throughout the past two decades. When individuals with exceptional expertise in one area work together with their counterparts in different domains, important developments are realized in all areas. Indeed, this Special Issue is filled with papers that are testimony to this process, bringing together the perspectives from many of those who were key pioneers, across all of these domains, in the development and application of this tool, discussing early developments and historical antecedents through to highlights of today's applications in basic neuroscience and medicine. To our community, the developments of 20 years ago were seminal—for some of us, they are the very foundation of our academic careers. Functional MRI has changed us and what we do. But as important as these developments were to the readers of this issue, another question can be asked: were these developments important beyond our small community? Has functional magnetic resonance imaging had an impact on the broader ways we practice medicine, do science, or live our lives? For better or worse, has fMRI changed the world?
This essay is based largely on a lecture given by the first author in May of 2011. We begin by describing his personal perspectives on the events of 20 years ago. The remainder of the essay examines the impact of fMRI on the broader stage of science, medicine, and society. As with any description arising from a singular point of view, there are inherent limitations—like a back projection reconstruction, the real image only forms from the sum of many perspectives. In the case of the present essay, this applies both to the history, and the selection of topics in the subsequent discussion.
The 10th anniversary meeting of the Society for Magnetic Resonance in Medicine (SMRM1) in 1991 featured two seminal presentations. The first was by Jack Belliveau, the winner of the society's Young Investigator Award that year, and the second was by Tom Brady, in a plenary lecture entitled, “Future Prospects for MR Imaging”.
Belliveau used dynamic susceptibility contrast MRI (DSC) to map changes in cerebral blood volume as a function of neural activation in response to a simple visual stimulus. The subtraction images he presented (Belliveau et al., 1991b), shown in Fig. 1a, represent the first unequivocal images of human brain activity changes seen using MR, and led to the iconic image featured on the cover of Science, shown in Fig. 1b.
Dynamic susceptibility contrast has long since been supplanted by BOLD-based imaging for most cognitive experiments (although it is still widely used for clinical imaging of perfusion and blood volume in the evaluation of stroke, brain tumors, and other cerebral pathologies). Nonetheless, it is worth noting that the discussion section of that poster's abstract (Belliveau et al., 1991a) included the observation that the potential presence of stable intravascular tracers, “offers the potential to perform continuous serial imaging of cortical function with subsecond temporal resolution… The ability to perform complete 2-D images in times as short as 50 ms implies that the hemodynamic response time of neuronal activation… will ultimately limit the temporal resolution of this technique” (Belliveau et al., 1991a). Jack Belliveau's work and enthusiasm brought brain mapping to our group in a way that inspired the entire laboratory, and Jack's prediction, above, did not have to wait long to be seen as prophetic.
These two ideas, dynamic imaging and endogenous contrast, were brought together in Ken Kwong's experiments of May 9, 1991 with human subjects. Ken's work was presented by Tom Brady in his keynote address at the same 1991 SMRM meeting. This was the first public presentation of Ken's work on dynamic BOLD and flow contrast functional mapping in humans using MRI. There were two endogenous contrast mechanisms used: arterial spin labeled flow contrast (Detre et al., 1992), and the intrinsic intravascular susceptibility contrast agent deoxyhemoglobin (Ogawa et al., 1990, Thulborn et al., 1982). Brady's presentation included a video of Ken's data, showing dynamic imaging of cerebral activation (Fig. 2). It was certainly not Hollywood quality, but many people in the audience appreciated its potential immediately; and for some, it changed the course of their careers.
It is inspiring to look at Ken Kwong's lab book from the day he first succeeded in this work. Two pages from this lab book, dated May 9, 1991, are shown in Fig. 3. On the left is a sketch of a “block design” for his experiment. Although the block design paradigm for fMRI seems obvious now, 20 years ago all functional brain imaging studies were done with single time-point injections, repeated many minutes apart, with analysis tools designed to identify the differences between sets of independent images. Ken's block design represented a new paradigm for activation studies, requiring different timing and analysis techniques, and remains the most commonly used functional paradigm more than 20 years later. The two sections on the right hand page show the contrast mechanisms involved. Ken first described an imaging sequence for BOLD contrast using gradient echo (GE) during the block design; then, immediately below, the T1-weighted inversion recovery (IR) imaging sequence to achieve contrast based on cerebral blood flow (CBF). Superimposed on these two pages is a visual activation subtraction map from the BOLD contrast experiments performed that day, and later published (Kwong et al., 1992).
At the time it was unknown whether BOLD, via gradient echo images, would yield SNR comparable to CBF-based imaging. Thus, on one day–May 9, 1991, Ken documented the conceptual paradigm for block-design fMRI and performed experiments using the two dominant modes of endogenous contrast that we use today: gradient echo BOLD imaging, and arterial spin labeling flow imaging. Not a bad day's work.
Beyond these and other early observations documented in this Special Issue, much has happened in our field. How can we think about the impact of this work? We will consider whether fMRI has changed the world by looking at its impact on each of the following domains: scientific impact, clinical practice, cognitive neuroscience, mental illness and society more broadly.
Section snippets
Scientific impact
One way to assess the impact of fMRI on scientific research is to use the tools associated with citation counts. We ask how fMRI compares with other high-profile scientific discoveries of the last few years. Table 1 illustrates one such comparison, based on the number of PubMed citations for eight high impact key words—words reflecting several recent Nobel prizes in medicine and physiology. On one hand, fMRI has not been cited nearly as many times as knockout mice, HIV or MRI more generally. On
Clinical practice
Perusing the 400 abstracts that were part of the Organization for Human Brain Mapping's first meeting in Paris (Mazoyer et al., 1995) reveals the vast range of medical topics for which fMRI was proposed very early on. Such topics included presurgical planning in conjunction with resection of the brain to treat tumors or epilepsy, treatment outcome evaluation after stroke, and diagnosis of psychiatric disorders, to name just a few. Have these early hopes been realized, after 20 years?
Cognitive neuroscience
Of course, when we think about functional imaging, we don't just think about pre-surgical planning and its clinical applications, we think about its ability to give us information and insights into the broader realms of human brain function. So what is the impact of fMRI in the domain of cognitive neuroscience? Here the data for fMRI's impact is more compelling—indeed, many of those 20,000 fMRI-related publications in Table 1 are directly or indirectly related to cognitive neuroscience. One way
Mental illness
Tom Insel, Director of the National Institute of Mental Health, when asked in an informal conversation: “What has been the impact of fMRI?” replied, “Despite 4000 papers, not a single finding has changed routine clinical care in psychiatry.” Pausing briefly, he then followed by saying that this will change in the next few years, and the change will revolve around a new definition of psychiatric diseases as disorders of brain circuitry—that this evolving understanding will change our definition
Society
Of course, every new technology also brings dangers. In 2002 The Economist featured a cover story about the ethics of brain science, in which they said, “Genetics may yet threaten privacy, kill autonomy, make society homogeneous and gut the concept of human nature. But neuroscience could do all of these things first” (Economist, 2002). And what was the focus of this dubious distinction? Functional MRI, of course, and the concern that fMRI would allow the creation of better lie detectors and
Closing remarks
The title of this essay is undeniably audacious. Each of the contributors to this Special Issue has their own views about the impact of functional MRI on their work, on their lives, and on the research community at large. Has fMRI changed the world? Table 2 summarizes the conclusions of the preceding discussion. Looking it over, the answer must be yes, and in ways that go beyond our laboratories. While there are always dangers, including potential abuses associated with mind-reading, lie
Disclosures
There are no relevant financial or relationship interests to disclose with regard to the subject matter of this presentation.
The following is the supplementary material related to this article.
Acknowledgments
I (BRR) am grateful to the International Society for Magnetic Resonance in Medicine for the invitation to give the 2011 Lauterbur Lecture, which was delivered at their annual meeting in Montreal, May 2011. This essay is largely based on that lecture. It was made possible through the privilege of working with a collection of exceptional people during the past 20 years, at the Massachusetts General Hospital and its Athinoula A. Martinos Center for Biomedical Imaging. I thank Judy Illes and Adena
References (43)
Perhaps functional neuroimaging has not told us anything about the mind (so far)
Cortex
(2006)- et al.
Functional magnetic resonance imaging (fMRI) ‘brain reading’: detecting and classifying distributed patterns of fMRI activity in human visual cortex
Neuroimage
(2003) - et al.
Neural substrates of choice selection in adults and adolescents: development of the ventrolateral prefrontal and anterior cingulate cortices
Neuropsychologia
(2007) - et al.
Images of psychopathology
Curr. Opin. Neurobiol.
(1998) - et al.
The neural bases of cognitive conflict and control in moral judgment
Neuron
(2004) - et al.
Neural predictors of purchases
Neuron
(2007) - et al.
Socioeconomic status predicts hemispheric specialisation of the left inferior frontal gyrus in young children
Neuroimage
(2008) - et al.
Oxygenation dependence of the transverse relaxation time of water protons in whole blood at high field
Biochim. Biophys. Acta
(1982) - et al.
Early ‘visual’ cortex activation correlates with superior verbal memory performance in the blind
Nat. Neurosci.
(2003) - et al.
Transcranial magnetic stimulation of the occipital pole interferes with verbal processing in blind subjects
Nat. Neurosci.
(2004)
Hominid brain evolution: testing climactic, ecological, and social competition models
Hum. Nat.
Language processing in the occipital cortex of congenitally blind adults
PNAS
Functional mapping of the human visual cortex by magnetic resonance imaging
Proc. Int. Soc. Magn. Reson. Med.
Functional mapping of the human visual cortex by magnetic resonance imaging
Science
Developmental differences in posterior mesofrontal cortex recruitment by risky rewards
J. Neurosci.
Perfusion imaging
Magn. Reson. Med.
Decoding mental states from brain activity in humans
Nat. Rev. Neurosci.
Functional imaging of intervention effects in stroke motor rehabilitation
Arch. Phys. Med. Rehabil.
Cited by (60)
Molecular basis underlying functional connectivity of fusiform gyrus subregions: A transcriptome-neuroimaging spatial correlation study
2022, CortexCitation Excerpt :It is widely accepted that a high degree of functional integration between distributed brain regions is necessary to support a specific neural process, e.g., visual cognition. Advances in functional brain imaging techniques, especially resting-state functional magnetic resonance imaging (rs-fMRI) (Logothetis et al., 2001; Rosen and Savoy, 2012), have enabled investigators to characterize functional integration of the brain by measuring the temporal coupling of neural activity across anatomically distinct regions, referred to as resting-state functional connectivity (rsFC) (Biswal et al., 1995; Yeo et al., 2011). Previous neuroimaging research has demonstrated that the FG can be parcellated into distinct subregions that present different anatomical and functional connectivity profiles (Zhang et al., 2016), endorsing the view that functional diversity of FG subregions may arise from differences in their connection patterns.
Quantitative assessment of visual pathway function in blind retinitis pigmentosa patients
2021, Clinical NeurophysiologyAnatomical and Functional Impacts of Congenital Bilateral Visual Deprivation on the Visual Pathway—A Comprehensive Review
2024, Journal of Clinical Medicine