The importance of neuroimaging techniques such as positron emission tomography (PET) and functional MRI (fMRI) in our understanding and treatment of mental illness is increasingly recognized. Neuroimaging studies have already helped refine the appropriate dosing of antipsychotics in schizophrenia (1); located targets for deep brain stimulation in refractory depression (2); predicted treatment response in obsessive-compulsive disorder (3); distinguished the neural targets of cognitive behavior therapy, dialectical behavioral therapy, and mindfulness meditation (4–6); identified target brain regions for abolishing substance dependence (7); and characterized pathological forms of social cognition in borderline personality (8).
Psychiatric residents typically receive little, if any, formal training in the principles and clinical uses of neuroimaging. Although one could argue that neuroimaging is of peripheral importance to current psychiatric practice, it will not remain so over the working lifetimes of today’s psychiatry trainees. At a bare minimum, they will need sufficient background knowledge to evaluate the merit (or irrelevance) of neuroimaging studies in the psychiatric literature. Future psychiatrists will also need to be able to use neuroimaging investigations effectively in patient treatment. In addition, only with a firm understanding of functional neuroanatomy will they be able to exploit the clinical implications of the new neurosciences of emotion, judgment, and social cognition. The time has come to provide psychiatric residents with a more formal curriculum in neuroimaging.
This article presents the format and curriculum of the University of Toronto Neuroimaging Week, a 5-day, intensive training course developed and overseen by a psychiatry fellow (OG) and psychiatric resident (JD). This program was specifically designed to prepare first-year psychiatric residents to use neuroimaging techniques in clinical practice, to evaluate neuroimaging studies on psychiatric issues (including pathophysiology, diagnosis, treatment, and outcome), and to understand basic psychiatric neuroanatomy, without which the neuroimaging literature is uninterpretable. We designed the program for first-year residents to integrate these core skills into the general medical training they receive in their internship year. Following a curriculum description, the paper presents objective and subjective measures demonstrating that Neuroimaging Week produces marked improvements in residents’ knowledge of, and attitudes toward, neuroimaging techniques in clinical psychiatry. Finally, this article suggests areas for future curriculum refinement and reviews possibilities for adapting Neuroimaging Week to other postgraduate programs or continuing medical education courses.
Neuroimaging Week was designed to provide intensive instruction in neuroimaging techniques and related topics. It was intended for trainees who had completed undergraduate medical training and were in their first year of postgraduate training. Internship-year trainees rotate independently through different clinical specialties with high workloads. To avoid major disruption of trainees’ clinical duties, the course was compressed into an intensive, dedicated week of study. During this period, trainees were relieved of usual clinical responsibilities aside from call duties. Creating this hiatus required reducing a 5-week clinical rotation block to 4 weeks, which was a challenge to arrange, given the many programs involved (e.g., emergency medicine, internal medicine). We collaborated with our program’s first-year coordinator to negotiate this reduction with the other programs affected.
Avoiding information overload was a major concern. In previous years, a Neuroscience Week course (with a broader curriculum covering diverse topics in neuroscience) had suffered a high attrition rate over 5 days of continuous, intensive instruction. Those trainees who did complete all 5 days reported difficulty assimilating new information by the end of the week, and few were able to complete the required readings.
To ensure that residents retained as much information as possible, Neuroimaging Week emphasized activities and interactive learning over pure lectures. The authors set a maximum of 50% lecture-format instruction for the course as a whole and (where possible) for any given hour of the curriculum. Demonstrations, discussions, question periods, tours, and activities filled the remaining time. Guest speakers were selected for rhetorical abilities and skill in audience engagement as much as technical expertise and content mastery. If speakers were unable to engage the audience, the course coordinators would intervene. Coordinators would insert deliberate pauses for reflection at regular intervals during instruction, to ensure that key concepts were understood and all questions were answered before approaching the next topic to encourage resident participation.
The curriculum avoided excessive focus on technical or theoretical minutiae and offered supplementary but no mandatory readings. From previous feedback, we predicted that this information would be poorly retained. In any case, many specific details would likely be obsolete by the time trainees entered practice. Instead, the emphasis of the course was on conveying clinically relevant principles concisely. These enduring principles were intended to build a basic foundation for ongoing learning over the trainees’ practice lifetimes, as the role of neuroimaging techniques in psychiatry becomes more prominent.
Neuroimaging Week assumed no formal background in functional neuroanatomy, slice-based neuroanatomy, or anatomy relevant to psychiatric disease. It also did not assume any prior knowledge of the fundamental principles of computed tomography (CT), MRI, fMRI, PET, or single photon emission computed tomography (SPECT), nor any formal training in interpreting CT, MRI, or other neuroimaging investigations. The course did assume some basic knowledge of neuroanatomy and neuroimaging techniques from undergraduate medical training, but no detailed knowledge beyond this level. Trainees in this program had reviewed some structural neuroanatomy during a dedicated Neuroanatomy Day 1 month prior to Neuroimaging Week, in which an anatomy program faculty member led an interactive demonstration employing brain models, anatomical specimens, and observed brain dissections in the Faculty of Medicine anatomy teaching laboratory.
The Neuroimaging Week curriculum covered four major themes. The first theme was psychiatric neuroanatomy: a review of structural neuroanatomy in two- and three-dimensional views; a review of functional neuroanatomy; and a survey of the emerging neuroanatomy of emotion, motivation, decision making, and social cognition—all common domains of pathology in psychiatric illness. The second theme was techniques and principles: a review of the principles behind CT, PET, SPECT, MRI, fMRI, and electroencephalography (EEG), with an emphasis on the practical clinical issues that arise in using each technique. Faculty from the departments of radiology and nuclear medicine collaborated in teaching these topics. The third theme was clinical skills. Topics included a systematic approach to interpreting CT and MRI images and the use of diagnostic imaging in neuropsychiatric case studies (taught in collaboration with neurology faculty). One example of a case study was multiple sclerosis presenting as first-episode psychosis (9). Sleep medicine specialists also lectured on EEG and conducted a tour of a clinical EEG laboratory. The final theme was literature appraisal: a systematic approach to the critique of a neuroimaging paper, a review of common problems with neuroimaging studies, and a guided, group-based critical evaluation of a neuroimaging publication in a prominent psychiatric journal (in this case, a PET study of the basis of pharmacotherapy and psychotherapy in treating depression (10)).
Figure 1 depicts the Neuroimaging Week schedule. Each presentation used interactive methods as well as, or instead of, didactic instruction: a brain-coloring activity for neuroanatomy, a patient-question activity for MRI principles, case discussions with neurology staff, sleep lab tours, open discussion of a neuroimaging publication, and a final review with teams competing to answer questions in a quiz-show format (Jeopardy!). Lectures on functional neuroanatomy of emotion, motivation, and self-cognition were open ended; the presenter solicited questions from the audience and then answered them with chalkboard illustrations, followed by a discussion. Virtually all other lectures used a semistructured format with frequent pauses for questions and answers, to maximize in-class comprehension and mastery of presented material.
Pre- and postcourse questionnaires used both subjective and objective measures to evaluate whether Neuroimaging Week achieved the desired learning objectives. Participation in this educational research study was completely voluntary and anonymous and had no bearing on trainees’ evaluations. The study was approved by the University of Toronto Research Ethics Board.
The precourse questionnaire surveyed trainees on age, gender, previous and planned electives in neuroscience or neurology, previous training in neuroimaging, and previous research experience. Subjective questions focused on overall expectations, personal learning goals, and personal comfort level with a range of curriculum topics. The postcourse questionnaire inquired about how well the overall expectations and personal learning goals for the course were met and surveyed trainees again on their comfort levels with the curriculum topics.
For objective measures of trainees’ preexisting knowledge base and subsequent knowledge acquisition, trainees completed a challenging multiple-choice questionnaire before and after the course. Twenty questions (∼1 question per 2 hours of instruction on a given topic) tested trainees’ knowledge of structural neuroanatomy, functional neuroanatomy, neuroimaging safety, MRI physics/functional MRI, nuclear medicine (PET and SPECT), and EEG/sleep medicine. Because the questionnaire was intended to evaluate the course rather than the residents, it was intentionally made difficult, to avoid ceiling or floor effects. To control for practice effects, the questionnaire came in two versions with equal levels of difficulty. Trainees were randomly assigned to two groups. Version A served as the pretest and version B as the posttest for group 1 and vice versa for group 2.
Thirty of 33 first-year residents agreed to participate in the study. Twenty-five trainees completed both questionnaires and were therefore included in subsequent analysis. Fourteen women and 11 men participated and ranged in age from 24 to 38 years old (mean age 28 years). Six participants held a graduate degree (chemistry, public health, pharmacology, toxicology, and history) in addition to a medical degree. Seventeen participants were currently involved in research (but only two in neuroscience research). No resident had received any prior formal training in neuroimaging.
Residents had high baseline enthusiasm. Sixteen of 30 residents expected the course would be beneficial (11 very beneficial) to their education; the remaining three had neutral expectations. For personal objectives, more than 90% wished to improve clinical skills in approaching and interpreting neuroimaging studies, knowing their general indications, and explaining them to patients. Most (∼80%) showed interest in interpreting functional neuroimaging studies or critiquing neuroimaging literature. The only area of limited interest (43%) was in understanding neuroimaging physics. After completing the course, residents reported that every surveyed learning objective had been well met on a Likert scale (1=strongly disagree; 5=strongly agree; mean=4.01, SD=0.14, range=3.72–4.16). Specific comments were overwhelmingly positive:
“Great examples of studies and their findings, very good explanations of tasks where certain areas of the brain are thought to be involved, very helpful that we were walked through the neuroimages on ‘approach to clinical MRI.’”
“Time for discussion, very interactive, very good use of analogies, the enthusiasm and good attitudes toward this topic were very inspiring.”
“Overall, an exceptionally well planned, delivered week of classes that I felt was very engaging and effective.”
Residents’ comfort with course topics also improved substantially (Figure 2). Precourse comfort ratings were universally negative. The lowest initial comfort levels were for PET in schizophrenia (1.40) and depression (1.37), interpreting neuroimaging/EEG studies (1.47–1.50), functional neuroanatomy (1.63), and literature appraisal (1.77). Postcourse comfort ratings were almost universally positive (mean=3.64, SD=0.40) and improved significantly in all areas (p<0.001, Bonferroni-corrected). The greatest gains were in literature appraisal (+2.07), functional neuroanatomy (+1.93), and PET in depression (+2.07) and schizophrenia (+1.92).
Twenty-five of 30 residents completed both multiple-choice questionnaires. Versions A and B showed no significant difference in overall difficulty based on questions correctly answered (A: 47.6%, SD=18.3; B: 49.7%, SD=14.5; t48=0.44; p=n.s.). Resident groups 1 and 2 showed no significant differences in overall performance across both tests (1: 46.6%, SD=17.7; 2: 50.6%, SD=15.0; t48=0.87; p=n.s.).
The multiple-choice questionnaires were intentionally made challenging to avoid ceiling or floor effects in performance. This goal was accomplished: pre- and postcourse tests showed low-moderate and high-moderate mean scores, respectively. Combining groups, overall scores were 36.7% (SD=11.5) on the precourse test, improving to 60.7% (SD=10.7) on the postcourse test (paired t24=11.28, p<10−10). All residents’ scores improved, with gains ranged from 6% to 45% (mean=24.0%, SD=10.6).
A growing number of psychiatry programs now offer longitudinal instruction in neuroanatomy and neuroimaging (10–12). Neuroimaging Week provides an alternative, intensive approach. To overcome the novelty and difficulty of the material, the course emphasized speakers’ rhetorical abilities, interactive learning methods, and mastery of enduring principles over specific details prone to obsolescence. This course was created not by senior staff but by a fellow and a resident with keen interests in teaching neuroscience.
Objective and subjective measures confirm the efficacy of this approach. Every surveyed resident showed a marked improvement in objective knowledge. Despite the difficulty of the questions, residents improved from 36% to 61% of the questions answered correctly. Practice effects cannot explain the improvement, because residents answered differed versions of the questionnaires before and after the course.
Subjective measures of engagement were similarly encouraging. Comfort levels improved from negative to positive for all but one topic surveyed. Notably, the attrition rate over the 5 days was much lower than anticipated from previous intensive courses. More than 25 of the original 30 residents completed Neuroimaging Week. Feedback was overwhelmingly enthusiastic, with many requesting further training in neuroimaging and neuroanatomy. None expressed fatigue or overall disappointment.
The Neuroimaging Week curriculum can be readily adapted to other psychiatry programs. Presentation can be longitudinal if necessary. A shorter, 2-day version could also cover many key topics: structural and functional neuroanatomy review, MRI principles and sequences, clinical cases, PET and SPECT principles, and guided literature review (Figure 1). This abbreviated version would also be suitable for continuing medical education.
Success of the curriculum depends on three factors: interactive methods to mitigate content overload, emphasis on key principles over technical minutiae, and rhetorical ability in the selected speakers. If speakers are not engaging, coordinators can intervene with questions and summaries to encourage participation. Programs lacking adequate infrastructure for demonstrations of PET, SPECT, and EEG techniques could invite speakers from neighboring institutions. Collaborators invited from neurology, radiology, and nuclear medicine provide helpful expertise. Practical activities, such as literature appraisal, serve a dual function: they help residents see clinical applications of neuroimaging and help coordinators assess residents’ mastery of the curriculum. Formal feedback ensures clinical relevance over successive iterations of the course.
Acquiring the background knowledge to apply neuroimaging research to clinical practice is not a straightforward enterprise, yet it is necessary for any resident who hopes to take full advantage of new discoveries in psychiatry, a field that is arguably evolving more rapidly than any other in medicine today. At a minimum, today’s psychiatric residents will need to be able to assess neuroimaging findings with a critical eye and determine if and how to apply them to their clinical work. More broadly, as neuroscience itself advances, psychiatrists of the future will need to detect the important implications for patient care. Neuroimaging Week, and other courses like it, can assist residents with this difficult enterprise. For these reasons, we favor adopting neuroimaging training as a core component of psychiatric residency training. With these skills, psychiatric residents will be able to translate the promise of 21st-century neuroscience into a new and more clinically powerful 21st-century psychiatry.
FIGURE 1. Neuroimaging Week Curriculum Timetable
Asterisks denote suggested topics for an abbreviated, 2-day version of the course.
FIGURE 2. Subjective Comfort Levels With Curriculum Topics Before (Black Boxes) and After (White Boxes) Completing The Course
Bars indicate 95% confidence intervals. (N=25)
At the time of submission, Drs. Downar, Krizova, and Zaretsky reported no competing interests. Dr. Ghaffar is a member of the editorial board of Cerebrio, a continuing medical education company, and receives honoraria from this company.