1: The Science,Art, and Philosophy of Radiographic Interpretation

The Science,Art, and Philosophy of Radiographic Interpretation

Matthew D. Winter

Department of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Florida, Gainesville, FL, USA


Diagnostic imaging is an art and a science. The science of diagnostic imaging is didactic information that is learned during veterinary training. The art is the experiential learning that takes place over the course of a lifetime as one interprets radiographs and the ability to extract information from an image. As one studies radiographs, one moves from a lower order of learning (pattern recognition) to a higher level of interpretation where different aspects of the interpretation process impact the final conclusions. Then, the interpretation is filtered through the clinical information relevant to the patient at hand.

Interpretation of a radiograph should be directed by a paradigm. An interpretation paradigm is a map that guides you along a path of thorough and complete evaluation of a radiograph. The paradigm is an essential tool to use for evaluation of all radiographic studies, and examples are provided in each section. This chapter presents an approach to and overview of the philosophies shared by the editors of this textbook regarding radiographic interpretation.

Why Radiographs?

Why bother with radiography, or diagnostic imaging in general? Specifically, radiography is relatively fast and readily available as a diagnostic imaging test. The procedures for most standard examinations are well defined, and expectations for the capabilities of the modality are relatively well understood. In the realm of diagnostic imaging tests, it is also inexpensive and noninvasive, a rapid test to perform. Therefore, it is a great tool to monitor and stage disease and evaluate anatomy. We also use radiographs to document the results of patient management, to figure out if a treatment is working or not. And of course, when possible, we use radiographs to actually diagnose disease.

​​Most often, imaging is used to decrease the level of uncertainty about a diagnosis. In most cases, we do not end up with a definitive diagnosis, but we do use the imaging findings as well as any other information to narrow the list of probable diseases. We do this by gathering data. If we think of each individual finding as a test, with each of those tests as having a particular value, then when we add them together, we should hopefully paint a picture or pattern of disease. So, think of each finding as a piece of a larger pattern. As we fit more and more of these diagnostic puzzle pieces together, the pattern becomes more clear. As that pattern emerges, hopefully we can recognize that it is consistent with a specific disease, or perhaps a subset of diseases. The result should be a shorter list of potential or probable pathologic processes (differential diagnoses), and we can direct our next steps accordingly.

Imaging Findings as Tests

As mentioned, tests can be characterized by their value, and that value is best described by sensitivity, specificity, and accuracy as tested against a normal and abnormal population of animals. If we think of each radiographic finding as a test, we realize that each finding can be associated with a certain number of true and false positives as well as true and false negatives when compared to a gold standard [1]. The gold standard would be the test considered to be the best available to diagnose a given disease. That said, the test may be relatively unavailable, too expensive, or perhaps very invasive, and therefore cannot always be done. The sensitivity or specificity for each and every finding for each and every disease are not always known, but we do have data for some of this, and we often can extrapolate. We also have our clinical experience and diagnostic acumen to draw on, which continue to grow over time. As more and more clinical research is done, we get new information on the value of findings as tests through science. Perhaps most importantly, the combination of findings can be most powerful as a diagnostic tool, and can further increase the sensitivity, specificity, and accuracy of radiography as a diagnostic tool for a given disease.

It is important to recognize that some individual findings may be very nonspecific, and that they are not exact for any particular disease and can be features of many different, completely unrelated diseases [1]. This means that, individually, they do not contribute to the reduction of uncertainty that we hope to attain. However, when we combine multiple findings, the added value of each finding narrows our scope in the “cone of certainty” (Figure 1.1).

Schematic illustration of the cone of certainty. A nonspecific change at the mouth of the funnel does not help narrow the list of differential diagnoses.

FIGURE 1.1 The “cone of certainty.” A nonspecific change at the mouth of the funnel does not help narrow the list of differential diagnoses. However, a series of findings added together improves our degree of certainty, narrowing the list of diagnoses at the tip of the cone. There are few diseases for which imaging findings are pathognomonic. However, a series of findings with varying degrees of value can result in a short and prioritized list of differential diagnoses that aid in decision making, clinical progress, and improved patient care.

For example, an unstructured interstitial pulmonary pattern that is moderate in severity and hilar in distribution could result in a large list of potential differential diagnostic considerations from multiple etiologies. If we combine this finding with other radiographic changes, such as left‐sided cardiomegaly, elevation of the carina on the lateral images, widening of the caudal bronchi on the ventrodorsal image and enlargement of the pulmonary veins in a dog, our differential list narrows very quickly to pulmonary edema secondary to left‐sided cardiac disease. Depending on other signalment and physical examination findings (small breed dog with a grade IV/V pansystolic cardiac murmur), our differential list narrows even more to mitral valve degenerative disease (endocardiosis) with secondary left heart failure.

This is why one of the most important and fundamental interpretation skills is learning to describe abnormal radiographic anatomy in an organized and systematic fashion. Being systematic and organized helps us to recognize patterns that might otherwise elude us.

Describing Abnormalities: Roentgen Signs

The fundamental language of radiographic interpretation is the Roentgen signs. These are the six features that we describe for every organ or body system that we evaluate: location, size, shape, number, margin, and opacity. The definitions as well as some terminology for use in description of abnormalities are listed in Table 1.1. Figure 1.2 is a radiograph that contains all radiographic opacities.

TABLE 1.1 Roentgen signs, definitions, and terminology.

Roentgen sign Definitions Abnormal descriptive terminology
Size The relative extent or dimensions of an organ or object on the image. This can be an absolute measurement in mm or cm, or may be a ratio formed by comparison to a standard (i.e., vertebrae, pelvic diameter). The description should always be relative to the expectation of normal for a given species and breed Enlarged
Increased in size
Reduced in size
Shape The external shape or contour of an organ or object. Most organs have a narrow range of normal shapes. Intestines are tubular, kidneys are, well, kidney shaped, etc. Round or rounded
Number A value representing quantity or amount. In its simplest form, we might identify that there are 2 kidneys, 7 lumbar vertebrae, or 10 pulmonary nodules. But we also might use this to characterize the specific quantity of cardiac chambers or liver lobes enlarged or affected by disease Value (i.e., 3 pulmonary nodules)
Increased in number (compared to normal or a prior study)
Decreased in number
Margin The edge or border of a structure or organ. Smooth
Normal or abnormal contour
Location Place or position. Most organs have a normal, expected position that can be altered by disease. In many cases, the position of an organ may be altered by an adjacent abnormality. Recognizing this is key to understanding the lesion. Knowledge of radiographic anatomy is of the utmost importance. Remember that “Anatomy is Power!” Normal
Displaced (dorsally, ventrally, laterally, to the left, etc.)
Opacity The relative ability to attenuate x‐rays. There are five radiographic opacities. Relative differences in the soft tissue opacity of organs are often related to physical density or thickness Gas, fat, soft tissue/fluid, mineral/bone, metal
Photo depicts postoperative lateral radiographic image of the right crus of a dog that contains all radiographic opacities.

FIGURE 1.2 Postoperative lateral radiographic image of the right crus of a dog that contains all radiographic opacities. Gas is evident outside the patient, but also notice the subcutaneous gas cranial to the femur and caudal to the distal tibia (arrowheads). Gas is also superimposed/within the musculature caudal to the crus (open arrowheads). Fat is present in the subcutaneous tissues (asterisk). The musculature of the limb is soft tissue opaque. The variable shades of soft tissue are related to thickness. The femur, tibia, fibula, tarsal and metatarsal bones are mineral opacity. The implants are metal opaque.


Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Apr 16, 2023 | Posted by in ANIMAL RADIOLOGY | Comments Off on 1: The Science,Art, and Philosophy of Radiographic Interpretation

Full access? Get Clinical Tree

Get Clinical Tree app for offline access