| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
1 From the Department of Radiology, Indiana University Medical Center, 702 Barnhill Dr, Room 1053, Indianapolis, IN 46202-5200. Received April 15, 2005; final version accepted April 29. Address correspondence to R.B.G. (e-mail: rbgunder{at}iupui.edu).
Educationally speaking, one of the greatest strengths of expertise is also one of its most important weaknesses of expertise. The expert is able to complete complex tasks quickly and apparently effortlessly. For example, a chess grand master can glance at the pieces on a chessboard and instantly see where the advantage lies and what moves should follow. A dog breeder can glance at a dog and instantly tell what breed it represents and how it is likely to fare in competition. Likewise, a competent radiologist can glance at a chest radiograph or head computed tomographic (CT) scan and establish an accurate differential diagnosis, frequently even honing in on the single correct diagnosis (1). This kind of expertise offers a great advantage, because it permits very effective and efficient practice. When it comes to sharing such expertise with novices, however, it can be problematic. The expert tends to take for granted things that the novice may be unable to appreciate, let alone understand.
If radiology educators are to excel at their craft, they need to better understand what learners know and do not know. When they point out features of diagnostic images, they tend to assume that learners can see what they are talking about. They make recommendations regarding the correct sequence of examinations in a diagnostic work-up that seem obvious to them but may be difficult for learners to follow. They tell medical students and residents what they most need to learn, but they may not adequately explain to them why they need to learn it. Separating the expert from the novice involves considering not only the content and quantity of knowledge but also the individual perspectives from which the expert and the novice approach the diagnostic task. The issue here is how to teach the logic of radiology—or rather how to make this logic available and accessible to novices.
If we were merely downloading or transferring information, then education would be easy. Because radiology educators are attempting to transform the novice's perspective, it is more difficult. Educators are so familiar with what they know that it is hard to recognize it. They take it for granted. How do expert radiologists interpret images? How do they determine what examination to order? How do they determine what to attempt to teach learners? These are different processes of reasoning—different logics. For education to be effective, radiology educators must learn to recognize these logics for what they are, share them with learners, and see how they fit together in the integrated practice of radiology. Do they conflict with one another or reinforce one another, and how can educators articulate them in ways that contribute to more effective patient care?
We misunderstand novices if we see them as miniexperts who are already looking at things the way experts see them and need to merely invest more time and effort to become experts themselves. Experts' memories play tricks on them, and over time they forget how difficult it was to gain a workable understanding of these logics. With time, they become so embedded in experts' everyday thoughts and actions that they no longer notice them. They are like fish that do not notice the water in which they are swimming. Experts seldom have time to reflect directly on the underlying principles that make their work possible. If we could gain a better understanding of what novices need to know and what is needed to help them develop it, we could do a better job of educating them. We need to find means of making experts' basic assumptions and routine thought processes explicit so that students can see more clearly what a radiologist does and thus do a better job of learning it.
What do radiologists really do? From the novice's point of view, radiology may resemble the act of shining a flashlight in a dark room. When you are looking for something, all you need to do is point the flashlight in the right direction and the answer will appear before you, full blown. When a patient presents with hemoptysis, one needs only to obtain a chest radiograph and inspect the image to determine whether or not the patient has lung cancer. Novices may liken the physician to a person in a dark room without a flashlight and see radiology as the flashlight that illuminates the diagnosis. Of course, this is a rather naive view of both the diagnostic process and the role of imaging technology. This view misrepresents the role of technology in mediating observation and interpretation (2). The referring physician is not completely in the dark, and radiology involves a great deal more than merely illuminating diagnoses. Yet learners may see this simplistic account as precisely the view of radiology that we promulgate.
Even when images reveal the answer, they never do so by themselves (3). If the clinical question concerns a skeletal fracture, radiographs of the affected bone usually reveal whether or not it is broken. The accuracy of the test, however, is not perfect. In some cases, other views or modalities may be necessary. To see a nondisplaced carpal scaphoid fracture, we may need to obtain special scaphoid views. To see demyelination in the deep white matter of the brain, we may need to order a magnetic resonance (MR) examination of the brain instead of CT.
Medical students tend to regard radiology as a window to human biology. For them, ordering a radiologic examination is like pulling back a curtain to see what lies behind it. When pressed to liken radiology to another medical specialty, they often point to pathology, a specialty that relies on a set of techniques (eg, dissection, microscopy, etc) that are as broad as radiology techniques, to reveal underlying biologic mechanisms (4). From the novice's point of view, radiology and pathology share the ability to render the invisible (eg, depth, structure, abnormalities, etc) visible. Anthropologist Michel Foucault saw the transition from a superficial focus to a deep focus as the essential moment in the birth of modern medicine (5).
Merely looking at a series of images is not enough. Educators cannot necessarily blame a learner's failure to detect a lesion on a lack of radiology experience. Nor is it sufficient to spend more time looking at the images. The novice's eye and mind must be trained to see anew—or at least to see in a new way. It is a mistake to explain learners' failures to detect pulmonary nodules or pneumothorax strictly in terms of failed memorization or effort. Educators should not cultivate in learners the impression that novices require only practice, practice, and more practice.
Most of what radiologists do is interpretive rather than elucidative. Radiologists do not merely pull back a curtain and reveal the clinical truth. Instead, they actively make sense of what they see. They are not merely revealing what is there. They are trying out different hypotheses in an effort to name the underlying abnormality. When educators stick to the language of science in explaining radiology to students, they run the risk of misleading them. Students and novices tend to have a rather underdeveloped understanding of what science is and how it works. They tend to equate hypothesis construction and validation with simple pattern matching. However, radiology involves more than merely flipping through a deck of mental images to see which one best fits the picture.
Educators' best attempts to explain what radiologists do are often lost on learners without the educators realizing it. Some radiologic information is strictly image based. For example, the radiologist attends to size, shape, borders, tissue characteristics, and enhancement when trying to classify a lesion. Yet other nonvisual forms of data may be equally or even more important. For example, what are the patient's symptoms? What findings are present at physical examination? What do the patient's laboratory results reveal? Are other imaging studies available that might show how the lesion is changing over time?
The nonimaging clinical context is crucial to understanding what images mean. A differential diagnosis framed in the context of a patient assumed to be afebrile may change substantially if it turns out the patient has a high fever. Lesion characteristics that point to one diagnosis in a patient who has been sick for months may direct the radiologist somewhere quite different if the patient actually became acutely ill within 1 day. The presence or absence of immune compromise or a prior history of malignancy may totally transform the diagnostic approach to an imaging examination. In some cases, it may even suggest that the wrong examination has been performed.
Educators need to do a better job of communicating to students that radiology is a fundamentally interpretive enterprise. Radiologists not only see but also think. Although radiologists often refer to themselves as reading images, they do not simply read images the way one might read the diagnosis from a patient's chart. If educators foster such misperceptions, they are likely to compromise their learners' performance.
If referring physicians think radiologists literally read biologic features from diagnostic images, then the quality of their consultations may be undermined compared with what it could be if they recognized the larger interpretive context in which radiology operates. Knowing that every image is interpreted against the background of clinical history, a savvy referring physician recognizes the value of providing pertinent clinical information. It is also important for referring physicians to understand that most radiologic interpretations are probabilistic. Diagnostic certainty is the exception rather than the rule, and radiologists usually think in terms of the more likely or less likely hypothesis. For example, they often remind themselves that uncommon presentations of common abnormalities are more common than common presentations of uncommon abnormalities.
Radiology involves at least two levels of interpretive activity: One is the interpretation of images, and this is what learners usually envision when they think of radiology. The second level is not image based. When radiologists try to determine whether an imaging examination is indicated, what kind of examination will be most helpful, how to perform the examination (eg, whether or not to administer intravenous contrast material), and how to make sense of the images, they are not merely interpreting images but also thinking like radiologists. The images provide a window to biologic mechanisms, but additional forms of interpretation are still necessary if imaging is to contribute optimally to patient care.
Radiology educators need to make certain that students understand that expert radiologists cannot gaze on biologic reality in an unmediated fashion. Perceptions and judgments are always shaped by the theoretic framework in which we approach them. If learners do not understand this, they are not really grasping radiology. To avoid this pitfall, it is important to invite learners to function as collaborators. Educators and learners need to "unpack" together what they mean by radiologic interpretation and the logic sets that support acts of interpretation in the discipline.
To bring the issues of interpretation and logic to the surface, educators should help students answer particular kinds of questions: How would you decide whether or not to request an imaging examination in this case? What examinations would you consider, and how would you choose which is best? Is diagnostic accuracy your sole selection criterion, or do other factors such as risk and cost enter into your deliberations? What clinical information would be important to you in interpreting the images? What question are you really trying to answer, and how will your answer be used in caring for the patient? Is there anything that radiologists could do to improve the consultative process for the patient or referring physician?
Experts do not have full image tanks; nor do novices have empty image tanks. A novice radiologist is someone who has limited knowledge of what radiologists do and how radiologists think. The goal in teaching them should be to provide an introduction to medicine from a radiologist's point of view. In fact, if educators were to reengineer what they teach medical students, then providing such an introduction would be the theme around which they would need to consider organizing the curriculum. It is not enough to make classes problem based. Educators need to explicitly highlight the principles that underlie the work they do as radiologists. What, for example, is radiology's epistemologic stance toward pathology? What counts as a clinical problem? What kinds of evidence and arguments do radiologists bring to bear on clinical cases?
What we are talking about here is unpacking what radiologists have taken for granted or have so deeply embedded in their thoughts and actions that it has become common sense. It is often difficult for experts to retrieve and recover this deep understanding without help (6,7). The recovery of common sense has been at the heart of the social sciences since the 19th century. If radiology educators choose to adopt this project, they may want to call on others for assistance. One strategy would be to see what can be learned from social scientists who have attempted to identify the first principles of medical work (8).
The goal is to unpack expert knowledge as radiologists understand the term. Educators need to let learners see as much of the essence of radiologic thinking as possible and encourage them to see their patients' care from this perspective. They should invite learners to work with them to develop shared understandings of particular cases. Their assessment of learner performance should involve more than simply asking "Do you see it or not?"
The crux of the matter is how expert radiologists deal with uncertainty. Learners tend to suppose that simply knowing the diagnosis with complete certainty is the ideal. This attitude is manifested in their naive belief that they will be able to establish a diagnosis for every case. They may also expect to be told right away whether they got it right or not. The appropriate goal of clinical practice, however, is not to know the diagnosis but rather to know what to say and do next. Choices can be limited and ranked. Learners can improve in avoiding choices that are likely to be unproductive or even harmful. In effect, educators are asking students to learn two things at the same time: to think globally and to think locally.
If educators understand what learners really need to know about radiology, they can use these needs as organizing principles for determining what to teach and how to teach it. They may end up focusing less on pure image interpretation and more on the larger clinical context in which radiology is situated and the role imaging plays in contemporary health care. If educators think of radiology education in these terms, they will be in a better position to help learners such as medical students begin to think like radiologists, understanding the distinct interpretive contributions radiology can make to their patients' care.
In summary, the ultimate goal of radiology is not to reveal findings; nor is it to recognize the findings that diagnostic images reveal. The ultimate goal of radiology—and therefore of radiology education—is to help patients. To do this, radiology educators need to develop and teach the best interpretive paradigm for the discipline that they can find. This requires them to pay careful attention to how they think and work each day. It is within this context—and this context alone—that educators will be able to provide a radiology education that most enriches students' understanding.
 |
References |
|---|
 TOP References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
