New Prospects and Ethical Challenges for Neuroimaging Within and Outside the Health Care System

In a recent AJNR editorial, Illes (1) described a new discipline of neuroethics emerging at the crossroads of biomedical ethics, research, and clinical neuroscience. With an explosion of studies and unprecedented applications of functional neuroimaging, especially with MR imaging (2), neuroimaging has garnered significant attention from the discipline. New capabilities have enabled functional mapping of complex human behaviors such as moral reasoning and racial stereotyping never before imaged in the research environment and have begun to provide new forms of quantitative data about neurologic disease that may lead to improved diagnosis and treatment and even to predictive markers of disease. In the past, however, technological advancements have often outpaced consideration of their ethical, legal, and social implications (3). Here we present an initial approach to conceptualizing the neuroethical considerations for advanced neuroimaging, with a focus on the practical translation of capabilities from the laboratory to the clinical environment and outside the health care setting.

In an analysis of the distribution of studies involving functional MR imaging (fMRI) with a clinical component conducted between 1991 and 2001, we found that, collapsed over time, presurgical mapping studies (ie, for tumors and epilepsy; 33% of clinical studies), major psychiatric disorders including depression (18% of the studies) and neurodegenerative diseases including Alzheimer disease (12% of the studies) accounted for 63% of the data. The remaining 37% of the studies were divided among other categories such as drug and alcohol use, nonpathologic changes over the lifespan, and developmental pediatric delays. We also found that of a total of 642 review articles, 74% were devoted to either clinical studies (52%) or methods development (22%).

We infer from these data that there is vigorous momentum to transition imaging capabilities from the research setting into practical application. We are compelled to ask concurrently what moral reasoning will be needed to determine the trade-offs of risk and benefit of such complex new information in the clinical environment. How will a visual activation image affect physician practice patterns or patient insurability? How will this new form of quantitative information be protected, and what impact might the evidence provided by a brain image have on a patient’s understanding about his or her own disorder? Such issues may be especially acute for disorders for which qualitative results from clinical or neuropsychological examination were the exclusive basis for diagnosis in the past, for functional images that are discordant with results obtained by using reference standards, and even for behaviors newly “medicalized” by imaging findings and not previously considered pathologic.

The issues for predictive imaging are no less trivial. Well known to the field of genetics is the ethical quandary of predicting the likelihood of a disease, such as Alzheimer disease, for which there is no cure at present. Whether neuroimaging comes to be used alone or adjunctively to genetic testing or others, neuroimagers will have to face old questions for the new domain: who should be tested; what procedures are needed to promote good surrogate decision making for impaired patients; what safeguards are needed for ensuring confidentiality, access to counseling, and protections from inappropriate advertising and marketing (4)?

In the public health arena, how will we manage brain activation information that might predict a propensity for sociopathy and suicide in adolescents and aggression in adults? If we project that neuroimaging services will become openly available in the consumer marketplace—like self-referred body scanning—they may also become available in our school systems. What are prospects for using neuroimages for screening or for justifying remedial training or therapeutic enhancement for behaviorally difficult students, those who have learning disabilities, or students who are gifted? Core issues such as who will have access to interventional programs, who will pay, and what is the duty to inform third parties engender significant moral debate. Further, but perhaps even more in the future, could advanced new medical capabilities such as those afforded by fetal MR imaging for the diagnosis of central nervous system disorders eventually become adopted for predictive screening for complex behavioral traits?

In an era of increasing violence in our society and increasingly powerful imaging capabilities for detecting neurobehavioral phenomena such as lying and deception (reviewed by Illes et al. [2]), the implications for responsible application of the technology in the criminal justice system also quickly surface. With heightened media attention to such scientific advancements and the predilection for juries to give great credence to expert testimony and evidence, appropriate dismantling of information available from visual images—whether they are structural CT studies, or any of an array of functional images including positron emission tomography, single photon emission tomography, electroencephalography, magnetoencephalography (MEG), or MR—by appropriately trained neuroimaging experts is critical to effective communication of the information that may be correlated to either guilt or innocence. In parallel, and as the ubiquity of neuroimaging technology such as fMRI becomes further realized, screening in highly trafficked public areas such as our national airports may become a true possibility. Who will be screened, who will interpret the data, and how the data will be used are but a few of the challenges with which we will be faced.

Once priorities for advancing neuroimaging capabilities are identified by neuroradiologists and others within the health care setting and outside it, a framework for addressing them will evolve through broad acceptance of the issues, a common language for engaging in dialogue about them, and evidence-based approaches to study them. New information available from brain images will also undoubtedly inform beliefs and practice involving mutual influences in the relationship between brain and behavior throughout the lifespan. Neuroethics will invariably be concerned with and call for further discussions of these issues. Whatever shape these discussions may take, it will be imperative to think about how to adjudicate between biologic influences on behavior in health and disease, environmental, and cultural effects and factors that are a function of the choices we make.