X-rays, like light, are part of the electromagnetic spectrum but have a higher energy and shorter wavelength, which allows them to pass into and through the human body (Fig. 14.2). Different tissue types absorb different amounts of the X-ray beam: bone, a very dense tissue, absorbs most of the beam whereas lungs, consisting mainly of air, do not. As a result, varying amounts of X-rays exit the body, reflecting the different tissue types that the X-ray beam has passed through. As Roentgen discovered, the X-rays will stimulate a fluorescent screen to produce light, which can be captured on specialized photographic film to produce an image of the body part. This is referred to as a radiograph or X-ray. Obtaining a good radiograph has many similarities to obtaining a good photograph. For example, if the subject moves, the photographic image will be blurred; if not enough light is available, the image will be dark. Similarly, if the radiograph is not exposed correctly, the image will be limited. This is particularly a problem with large patients. These issues are the same for all imaging studies.

Radiographs are relatively inexpensive, quick to perform and widely available. As a result, they continue to make up the bulk of imaging studies. Similar to the camera, radiographic images have moved away from film and are now mainly of digital format.

Image quality depends on the ability to discriminate between two adjacent objects, also called spatial resolution; with X-rays, this is very good. However, the ability to see a structure also depends on the difference in the attenuation of the X-ray beam between different tissues, i.e. tissue contrast. For example, bone, which attenuates most of the beam, is well outlined against adjacent soft tissue such as muscle, which does not attenuate the beam as much. Similarly, the lung parenchyma and cardiac silhouette is differentiated well from the surrounding air (Fig. 14.3A). However, differentiating soft tissue is a problem because of a similar attenuation of the beam; this means that the tissues all have a similar gray appearance, i.e. little or no contrast (Fig. 14.3A). This can be overcome to some extent by using contrast agents. Oral contrast with a barium solution, e.g. a barium enema, is used for evaluating the gastrointestinal tract. Intravenous and intra-arterial iodinated contrast agents for evaluating blood vessels and organs are used in arteriography and intravenous urograms.

Standard X-ray images produce a two-dimensional display of a three-dimensional structure; this means that there is an overlap of structures resulting in part of the anatomy being hidden. For example, on a frontal chest radiograph, the overlying heart obscures the spine (compare the chest radiograph in Fig. 14.3A with the chest CT in Fig. 14.3B and C). This can be alleviated somewhat by obtaining additional views such as a lateral view. However, the problem of overlapping structures has been overcome following the introduction of computerized axial tomography (CAT or CT scan) in the 1970s, made possible with the advent of the computer and the ability to capture and store data digitally.

A computed tomography (CT) scan also utilizes X-rays to produce images but, instead of being stationary, the X-ray tube rotates around the patient (Fig. 14.4). Specialized detectors collect the emerging X-rays and produce an electrical signal that is fed into a computer; this information is then used to construct a cross-sectional image (Fig. 14.3B). The initial CT technique took several hours to acquire the data and several days for the computer to reconstruct the images. Modern scanners are very fast and can image the entire abdomen and pelvis in seconds. Multislice CT scanners are now replacing single-slice technology and, as the name indicates, multiple slices can be produced per rotation of the X-ray tube. This has the advantage of covering more territory and shortening scan times. This technology can produce very thin axial slices allowing the production of isotropic data sets, i.e. it has the ability to acquire a volume of data that can be viewed in any plane without distorting the image. Until this innovation, CT scans were generally limited to an axial display; now images can routinely be displayed in any plane giving multiplanar (MPR) and three-dimensional reconstruction capabilities. This allows more information to be gleaned from the study and improves diagnostic accuracy (Figs 14.5 and 14.6), particularly with complicated pathology such as complex fractures (see Fig. 14.7).

CT scanning is used to image all parts of the body. The speed of the examination makes this a particularly attractive investigation for older patients who may find it difficult to lie still for long periods of time. It also makes it the preferred method for examining very sick and injured patients when time is of the essence (Figs 14.5 and 14.6). The relatively large size of the gantry (the part of the machine that the patient passes through) (see Fig. 14.4) has almost eliminated the problem of claustrophobia. As with standard X-rays, CT demonstrates superb lung and bone detail (Figs 14.3C and 14.7). Fluids, such as ascites and cystic structures are well defined as are calcifications and acute and subacute hemorrhage (see Fig. 14.12A–C). Fat, particularly in the abdomen, is a natural contrast agent and is useful in defining adjacent organs and inflammation (see Fig. 14.6); however, CT still requires additional contrast agents to improve visualization. Oral contrast, usually in the form of dilute barium, aids visualization of the bowel (see Fig. 14.6), and intravenous contrast is used to optimize the evaluation of veins and arteries and the vascularity of organs. The addition of intravenous contrast material also allows further uses of CT, for example pulmonary embolism is now routinely evaluated by this method (see Fig. 14.5). The speed of the new multislice scanners allows the beating heart and coronary arteries to be evaluated. The ability of CT to demonstrate an artery is now as good as the more invasive procedure of diagnostic arteriography and, in complex anatomical situations, may be better. As a result, many arterial lesions, such as aneurysm, dissection and stenosis, are diagnosed and evaluated with CT (Fig. 14.8), and the more invasive diagnostic arteriogram is now mainly used for therapy (e.g. treatment of a narrowed artery and aortic aneurysm with specially designed stents, which can be placed percutaneously via adjacent non-diseased vessels).

Ultrasound has been used since the 1950s to produce medical images using sound waves. The frequency of the wave used to produce the images is in the range of 1–20 MHz. It is called ultrasound because this frequency is above the human audible range of 2–20 kHz. The probe or transducer used during the examination is responsible for both producing the sound and collecting the sound waves reflected from the patient’s organs and tissues. The received sound is converted to an electrical signal and, from this, an image is developed by a computer and displayed in real time on a TV monitor. In addition, by using the Doppler effect, the returning sound waves can be used to evaluate blood flow. It is excellent for imaging soft tissues and has a long list of uses, including imaging of the major abdominal and pelvic organs. Superficial structures are particularly well seen, for example, the thyroid, superficial tendons, muscles, veins and arteries (see Figs 14.14C and 14.15E and F).

The phenomenon of magnetic resonance was discovered in the 1930s and initially used to determine the composition of chemical compounds. In the 1970s, it was realized that the same techniques could be used in medical imaging and, by the 1980s, magnetic resonance imaging (MRI) units were in clinical use. Magnetic resonance uses radio waves and a strong magnetic field to produce the image; as with ultrasound, it does not use ionizing radiation. The technique relies on the fact that some atomic nuclei have magnetic properties that act like microscopic magnets when placed in a strong magnetic field. The human body has an abundance of these in the form of the hydrogen ion that makes up water. These align with the direction of the bore of the magnet when the patient lies in the MRI machine (Fig. 14.9). If these nuclei are stimulated by a radio wave at a specific frequency, known as the resonant frequency, they gain energy and move into a transverse plane, perpendicular to the main magnetic field. This results in a radio wave being emitted by the rotated nuclei, which allows their position to be recognized. The radio waves are emitted and received by a device called a radio frequency (RF) coil, analogous to the ultrasound transducer that transmits and receives the sound waves. The coils are placed close to the patient. The head coil, used to image the brain, looks like a cylindrical cage that surrounds the patient’s head (Fig. 14.9A). As with ultrasound, the received signal, in this case radio waves, creates an electrical signal, which is fed into a computer to produce an image (Fig. 14.10A, B).

The high tissue contrast (i.e. the difference in signal between tissue types) afforded by magnetic resonance is responsible for the excellent depiction of soft tissue anatomy. There are two main types of pulse sequences used in MRI, which produce images of the same area but with a different contrast; these are known as T1- and T2-weighted images. This is achieved by varying the time when the RF pulse is emitted and when the returning RF wave is received. Fluid gives a low signal on T1 images and a bright signal on T2 images, whereas fat gives a high signal on T1 and T2 images (when using the faster T2 turbo spin echo [TSE] sequence). Unique to MRI is the ability to selectively remove or null particular tissues from the image. For example, fat, which is bright and may obscure pathology, can be removed by a technique called fat saturation or ‘fat sat’ for short (see Figs 14.13E and 14.15D). This is an extremely useful technique, which also makes it possible to confirm that a structure is fat-containing (e.g. a lipoma) by obtaining images before and after fat saturation.

MRI is also routinely used to evaluate blood vessels (see Fig. 14.8). These images can be generated without the use of intravenous contrast agents although contrast-enhanced studies are also used to acquire additional information. The latest and faster MRI sequences allow routine evaluation of the beating heart and are a valuable tool in complementing cardiac ultrasound and cardiac nuclear medicine studies. Magnetic resonance spectroscopy has the ability to evaluate the chemical composition of tissue and has shown promise in the diagnosis of cerebral tumors. MRI is also used to evaluate brain function, the so-called functional MRI.

There are some fundamental differences between nuclear medicine and the other imaging modalities. Whereas the other modalities rely mainly on a change in anatomy caused by a pathological process, nuclear medicine is able to show images of changes in function or physiology as a result of pathological change. The energy source in nuclear medicine is a radionuclide that emits ionizing radiation in the form of gamma rays, which come from the same part of the electromagnetic spectrum as X-rays. The radionuclides are tagged with a biological compound that is used by a living tissue; this combination is called a radiopharmaceutical. Unlike the other forms of imaging, the radiopharmaceutical is placed inside the patient, usually by an intravenous route, and taken up by the organ/cells or pathological process of interest. During decay of the radionuclide, gamma rays are emitted and pass out of the body and are collected by a gamma camera to produce an image. The most common radionuclide used is technetium 99 m, and a common study is a bone scan in which the technetium is labeled with diphosphonate (Tc-MDP) (Fig. 14.11A, B). This is quickly taken up by bone, particularly in areas of bone remodeling, for example fracture repair and most bone metastases. The camera is positioned over the area of interest and images obtained in a two-dimensional plane (the planar image) (Fig. 14.11C). As in CT, images can also be obtained by rotating the camera slowly around the patient to obtain a cross-section or tomogram. This is referred to as SPECT (single photon emission tomography), and is used in, for example, SPECT bone imaging and thallium SPECT imaging of the heart.