Issues of Concern
In nuclear medicine, imaging artifacts can be broadly divided into instrumental, radiopharmaceutical, technical, patient, and treatment-related.[2]
Instrumental
With digitized gamma cameras, the acquired image raw data may show artifacts clearly, but post-processing, the artifacts may become subtle or altered and difficult to recognize. These artifacts can be categorized into uniformity, resolution and linearity, multiple window spatial registration, collimators, the field of view, and computer-related.[3]
- Uniformity artifacts can arise from faults in the photomultiplier tube, sodium iodide crystal, as well as other electronic and mechanical problems.
- Resolution and linearity artifacts are principally affected by spatial distortion, but modern developments have given us automatic updating of linear correction maps with minimal human involvement for correction, even with PET/MRI using flood histograms.[4]
- Multiple window spatial recognition is the accuracy of dual-isotope positioning with multiple photons of different energy using their normalized reference ranges. Artifacts can occur when using multipeak radionucleotides such as 67Ga, 201Tl, and 111In, causing resolution degradation, causing subtle errors even with software corrections.[5]
- Attenuation artifacts can be inherent to gamma radiation due to the energy it has. Collimators address the vast majority of gamma camera attenuation potential problems, but Compton scatter artifacts can still occur.[6] Less energy causes more scatter and more absorbed radiation by the patient. Certain radioisotopes are used to minimize this effect (99mTc instead of 201Tl). Tomographic imaging (PET/CT and SPECT) reconstruction can avoid these artifacts by digital subtraction methods and three-dimension imaging. It does not change spatial resolution or radiation scatter content but increases contrast and helps with locational relationships of structures. Collimator artifacts have no bright rim around the photogenic area, which distinguishes them from crystal defects such as cracks. If the damage is more widespread, flood images can be compared between different collimators for detection. A dent or defect in the collimator can be caused during the manufacturing process or by mechanical stress. Software processing of the image can also reduce collimator artifacts.[7]
- Field of view artifacts can involve many mechanisms. Electronic masking of the unusable edge of view prevents edge packing effects along the crystal area. This can reduce the field of view by overcompensation or misalignment of the field of view (leaving edge effects in view). These artifacts can be prevented by correct calibration after servicing the machine. With any series of static images, misalignment may be a source of error. If multiple passes are used in acquisition or if a continuous acquisition is used, a “zipper” (linear) artifact or banding (often horizontal linear) artifact may be observed, respectively. This can be prevented by equipment maintenance of whole-body uniformity. Hard copies of images should be avoided as artifacts such as those related to the printing process and damage to the print medium are common. Truncation errors can also occur when the field of view between the PET and CT differ, positioning patients away from the in-plane center, or when patient size is too large for the field of view—even though this can mostly be corrected by software.[8]
- Computers are now integrated into all aspects of modern imaging, including nuclear medicine. This allows for a wide range of artifacts. Some examples include energy corruption, linearity and uniformity correction maps, failure in any step of digitization and image processing, or failure to display the image accurately.
Multimodality imaging is common in modern nuclear medicine. PET/CT and SPECT artifacts may include high attenuation bodies, truncation, respiration motion, and misregistration. They also include native CT artifacts, such as CT noise and CT slice thickness.[9] CT noise occurs with larger patients or in lower dose CT scans, exaggerated in the reconstruction. CT slices are too thick. It can create a loss of visualizing smaller details.
Radiopharmaceutical
Altered radiopharmaceutical biodistribution can be classified into two main categories: preparation and formulation and administration technique and procedures.[10]
There are various factors involved with the radionucleotide purity, including the radionucleotide itself (ex. isomer competition, radionucleotide contamination from cyclotron), its components (ex. reagent concentration, particulate size, commercial source variation), preparation procedures (ex. mixing order, incubation, competing components), and many others (ex. blood pH variations, blood electrolyte concentration, aluminum needle interaction, volatility, decomposition, etc.).[10]
Administration technique and procedure artifacts can be classified into five groups: extravascular injection, clotting in the syringe, suboptimal delivery in ventilation studies, administration line residue, and patient positioning.[10] Extravasation is common and is further discussed in the section of technical artifacts below.
Multiple-use of radionucleotides and study timing artifacts are rare but are something to consider with busy nuclear medicine departments where timing overlap may occur. The physical and biological half-lives must always be considered when more than one nuclear medicine scan is planned. Adequate clearance and decay time minimize the occurrence of such artifacts. Careful notice to radioisotope therapy timing with scintigraphy can also avoid occurrences. Biologic half-lives can be altered by patient actions such as having unscheduled meals (when fasting) or gut motility altering medications during gastric emptying or colonic transit studies. Food or medications may contain radiocontrast material and should be avoided during certain studies.
Technical
Injection site artifacts are usually avoided by software or lead blocking. This helps identify the site and minimize pixel overflow. Extravasation artifacts are common, and the artifacts include local accumulation and possible lymphatic tracking to regional lymph nodes to give false-positive nodes. The area of blood pooling can also cause Compton scatter effects by adjacent soft tissues, giving a false positive uptake.[11] Unintentional intra-arterial injections can create blood pooling or flow artifacts as well as delayed image findings.
Artifacts may also be caused by medical catheters and tubes inserted into the body. With venous catheter use, the radiopharmaceutical may adhere to the plastic wall, or residual amounts may remain in the catheter lumen or venous system if numerous valves present due to improper flushing with saline after administration. Pedal veins can be used to reduce the venous retention artifact if the radiotracer commonly does this (ex. 201Tl, (99m)Tc-sestamibi). Urinary catheter bags or nephrostomy bags can also impair views of the area if improperly positioned.
Patient-related
Anatomical variations such as skull thickness, muscle density, and adipose distribution may affect attenuation. Non-anatomical structures in planar imaging that can cause attenuation artifacts can be external (ex. clothing, coins, keys, belt buckle) or internal (ex. pacemaker, orthopedic devices, previously ingested contrast, breast implants, tampon). If the cause of the artifact cannot be removed (ex. limb plaster cast), then the reporting physician must be informed. Contamination artifacts are the most commonly found artifacts in nuclear medicine. These artifacts usually occur when radiotracer is found in unexpected areas.
Urine contaminating clothing or footwear frequently occurs. It can be confirmed and corrected by washing the skin or removing the contaminated clothing or different image views—lateral makes it obvious. The high photon concentration can impair the hip and pelvis visualization in planar imaging with incomplete voiding of the bladder. Bladder catheterization and reducing the time between voiding and imaging help minimize such artifacts. Secretions can contain radioisotopes, and this includes residual nasal and lacrimal secretions on handkerchiefs, skin, clothing but also may show interference surrounding ocular prosthesis. These can be interpreted as false positives.
Motion artifacts can exist in any PET or other radiological scan. Respiratory motion artifacts causing blurring, attenuation errors, false negatives (often near diaphragm), and possibly misregistration errors.[12] Misregistration usually occurs near the boundaries of organs. An example of a misregistration error is if a liver lesion is near the lung-diaphragm interface may show the lesion in the lung rather than the liver due to the respiratory artifact.[13] Respiratory gated (four-dimensional) PET/CT developments include hardware gating, software gating, Bayesian penalized likelihood (BPL) PET reconstruction, and texture analysis that aim to overcome inherent respiratory motion artifact.[14]
PET/MRI combinations are being used more frequently now but still have limitations due to breathing and motion artifacts. In the past, the most common SPECT imaging artifact was of the myocardium: the left breast or left hemidiaphragm, if the position changes between rest and stress imaging. This position change may be misinterpreted as pericardial effusion or false ischemic areas of the myocardium. Simultaneous or sequential image acquisition of transmission and emission is often used to correct attenuation artifacts, along with first order and extended acquisition correction techniques. Recent developments with multi-pinhole cadmium zinc telluride (CZT) gamma cameras, new detector materials, iterative reconstruction, and new isotopes (18F-NaF) help minimize artifacts.[15] Despite the advances in myocardial perfusion imaging, patient motion artifacts still exist mainly as coughing, twisting, sliding, and slumping during the acquisition.[16]
Treatment-related
Medical procedures such as radiotherapy affect the tissue by increasing uptake in the inflammatory stage and decrease uptake in the fibrotic stage with many radiopharmaceuticals. An example of this can be seen with treatment evaluation of bone metastasis if done too early. The post-treatment bone is damaged and remodeled, causing high uptake and false-positive progression for the metastasis when the bone is simply healing. This is sometimes referred to as a “flare phenomenon” in FDG PET.
Recent surgery certainly causes increase uptake as there is also inflammation and healing occurring. Subcutaneous injection sites may have increased uptake of certain radioisotopes (ex. 99mTc-MDP), and they may obscure regions (lumbar spine, pelvis, abdomen) on anterior planar views and can be confirmed with lateral planar or SPECT. Patients on dialysis and with chronic renal failure can have increased clearance through the liver to compensate for renal impairment, altering the biodistribution of the radiopharmaceutical.[17] These patients can also have various chemical interactions with the radiopharmaceuticals involving factors such as pH changes, electrolyte abnormalities, and dialyzing before sufficient target uptake.
Pathophysiologic and biochemical changes alter the biodistribution of a radiopharmaceutical. Examples include excretory organ failure, altered glucose metabolism (fasting on FDG myocardial imaging), abnormal hormone levels (hypoparathyroidism with the bone scan), inflammatory reactions, and numerous others.[17]
Therapeutic drugs the patient is taking may alter the biodistribution or pharmacokinetics of the radiopharmaceutical. An example could be cardiac medications interfering with stress tests by preventing the heart from reaching the desired rate. Another example would be immunosuppressants reducing chemotaxis, diminishing uptake of radiolabelled leukocytes. Some medications such as somatostatin analogs may alter the uptake of 68Ga-DOTATATE imaging.[18]
Clinical Significance
Distinguishing between variant (normal or abnormal) and artifact can be challenging. Knowledge and recognition allow for identification and possible correction to avoid errors in interpreting an imaging study. Along with the above categories (instrumental, technical, radiopharmaceutical, patient, and treatment-related), variants and the radionucleotide study used must be considered in the correct contexts.
Awareness of the potential artifacts is crucial to identify them.
Nursing, Allied Health, and Interprofessional Team Interventions
Both medical and non-medical personal are responsible for the quality of a nuclear medicine service. Although quality can be defined in many ways, using a myriad of selected indicators definitely involves interprofessional teams. Sometimes, patients' perceptions of quality and satisfaction may not agree with some of the measurements a nuclear medicine department uses. We can even find older patients are generally more satisfied than younger ones.[19]
Quality of service involves expertise from physicists, radio-pharmacists, nuclear medicine instrumentation, nuclear medicine technologists/radiographers, allied health professionals, nurses, and doctors. Artifacts in nuclear imaging can be identified in many ways, not just by the reporting specialist physician but also by other staff involved. Typically the reporting specialist physicians rely on staff members for updated patient and equipment information, but anybody in the imaging process can improve outcomes. For example, the person who transported the patient to the imaging department may report incontinence or that the patient is likely to cause a motion artifact if they are agitated so that adjustments might be performed for better imaging.
Compliance with guidelines and protocols and good communication between medical teams makes for the best outcomes, and being able to discern artifacts from images enables better treatment. The more relevant information provided and the better the communication between all teams will improve patient outcomes. If someone notices an anomaly and reports it to the correct staff member, corrections can be made to avoid any issues and improve patient care. Ideally, artifacts should be prevented and, if they are not, should be identified at the time of imaging or reporting.
Nuclear Medicine Artifacts