Ultrasonography

Ultrasonography is a medical imaging technique that uses high-frequency sound waves and their returning echoes to depict structures inside the body. A transducer transmits acoustic pulses and detects reflected signals, which are processed into images. Unlike imaging based on X-rays, ultrasonography does not expose patients to ionizing radiation. Its real-time images can show organ movement and blood flow, supporting diagnosis and image-guided procedures. (fda.gov)

Ultrasonography is used across many areas of medicine, including the assessment of pregnancy, the heart, blood vessels, abdominal organs, and superficial tissues. Beyond conventional anatomical imaging, Doppler techniques assess motion and flow, while elastography evaluates tissue stiffness. Examinations may use probes on the skin or specialized probes placed within body cavities or vessels. Image quality and interpretation depend on acoustic access, equipment, and operator expertise; diagnostic ultrasound also requires attention to acoustic exposure and biological effects. (nibib.nih.gov)

Acoustic Principles and Echo Formation

Ultrasound is sound with a frequency above approximately 20 kHz, the conventional upper boundary of human hearing; diagnostic imaging uses much higher frequencies, generally in the megahertz range. In soft tissues, these waves propagate principally as longitudinal disturbances: alternating compression and rarefaction produce changes in pressure as particles oscillate along the direction of propagation. Unlike X-rays, ultrasound is mechanical rather than electromagnetic radiation and requires a material medium. Frequency $f$, propagation speed $c$, and wavelength $\lambda$ are related by $\lambda=c/f$. Using the conventional soft-tissue speed of $1540\ \mathrm{m\,s^{-1}}$, a 5 MHz wave has a wavelength of approximately 0.31 mm. Higher frequencies provide shorter wavelengths, but their greater attenuation limits penetration. (nibib.nih.gov)

Pulse–echo imaging alternates transmission of short acoustic pulses with reception of returning echoes. A transducer emits a pulse into tissue; portions of its energy return from interfaces and internal structures, while the remainder continues deeper. Received echoes generate electrical signals whose timing and amplitude supply information about location and acoustic reflectivity. For a reflector at depth $d$, the round-trip travel time $t$ gives $d=ct/2$, with the factor of two accounting for outward and return travel. At $1540\ \mathrm{m\,s^{-1}}$, an echo returning after approximately $65\ \mu\mathrm{s}$ corresponds to a depth of 5 cm. This calculation assumes the propagation speed used by the scanner: actual differences between tissue speeds can produce errors in displayed distances. (nibib.nih.gov)

Reflection at a boundary depends on the difference in acoustic impedance between adjoining materials. For a plane travelling wave, $Z=\rho c$, where $\rho$ is density and $c$ is sound speed. In an idealized, lossless case at perpendicular incidence, the reflected fraction of incident intensity is

Equal impedances therefore produce no reflection in this model, whereas a larger mismatch produces a stronger reflected component. Boundary geometry also matters: a relatively smooth interface behaves as a specular reflector, so an oblique surface can direct reflected energy away from the receiving transducer. Differences in propagation speed can additionally cause refraction of the transmitted wave. (aapm.org)

Scattering redistributes acoustic energy in multiple directions when waves encounter small structures or spatial variations in tissue properties. The portion returning toward the transducer—backscatter—supplies echoes from within organs, rather than only from their larger boundaries. Echoes from numerous unresolved scatterers interfere constructively and destructively, producing the granular pattern called speckle. Absorption is distinct: it converts acoustic energy into heat through dissipative processes within the medium. Attenuation describes the reduction of wave amplitude or intensity during propagation, incorporating absorption and scattering losses; consequently, the echo available for detection depends on both the reflecting structure and the tissues traversed along the outward and return paths. (aapm.org)

A common approximation for soft-tissue attenuation is $0.5\ \mathrm{dB\,cm^{-1}\,MHz^{-1}}$, although values vary with tissue and frequency. Under this approximation, a 5 MHz pulse reflected at 2 cm depth traverses 4 cm in total and incurs approximately 10 dB of propagation loss, separate from the loss associated with incomplete reflection. This frequency–depth dependence constrains the signals available for reconstruction. Signal processing addresses how those signals are combined and interpreted, including reconstruction, noise reduction, and extraction of tissue information from echo statistics. NIBIB’s diagnostic-ultrasound research program explicitly includes innovative signal-processing methods, image enhancement, acquisition, and data-presentation techniques. (aapm.org)

Transducers, Scanner Architecture, and Examination Routes

The ultrasound transducer converts electrical excitation into acoustic pulses and returning echoes into electrical signals. In conventional probes, its active elements are usually piezoelectric ceramics: an applied electric field produces mechanical deformation, while mechanical stress from an incoming sound wave generates an electrical response. This reversible piezoelectric effect allows the same elements to transmit and receive during successive phases of a pulse–echo acquisition. The probe is therefore both an acoustic source and a detector; the scanner supplies the excitation and processes the much weaker echo signals. (nibib.nih.gov)

Array transducers contain multiple elements whose excitation timing can be controlled individually or in groups. Relative delays focus the transmitted wave and, where the array geometry permits, steer it electronically. During reception, a beamformer delays and combines signals from different elements so that echoes from a selected location add coherently. Two-dimensional matrix arrays extend electronic control across both dimensions of the aperture, but increase the number of connections and acquisition channels. For example, a fully connected $32 \times 32$ array contains 1,024 elements and would require 1,024 channels if every element had a dedicated channel. A 2024 experimental study of hybrid beamforming examined grouping channels into analog subarrays before digital combination to reduce this hardware burden. (ti.com)

A typical digital scanner separates transmission, acquisition, signal processing, and display functions. High-voltage pulsers excite the array through transmit–receive switches, which protect sensitive receiver circuits during transmission. Returning signals pass through low-noise amplification, variable gain, filtering, and analog-to-digital conversion. Time-gain compensation increases amplification for later-arriving echoes to counter the greater attenuation associated with longer tissue paths. Receive beamforming combines the digitized channels; subsequent processing detects echo envelopes and compresses their amplitude range. Scan conversion maps the acquired scan-line coordinates onto the display’s pixel grid. A Texas Instruments architecture paper published in November 2008 documents this chain, including the controller that synchronizes excitation and acquisition, processing hardware, memory interfaces, and the display. (ti.com)

For external examinations, the operator places a probe on the skin and moves or tilts it to obtain different views. A thin layer of coupling gel fills gaps between the probe face and skin, excluding air pockets that would obstruct acoustic transmission. The gel is an acoustic coupling medium, not an injected contrast agent. External scanning leaves the transducer outside the body; even with good surface contact, the available view depends on the tissue path between the probe and target. (fda.gov)

Internal examination routes bring the transducer closer to structures that may be less accessible from the body surface. A transvaginal probe is inserted into the vagina to obtain views of the uterus and ovaries from different angles, rather than transmitting through the anterior abdominal wall. In endoscopic ultrasound, a transducer is incorporated into an endoscope tip or introduced through its instrument channel. This arrangement images the gastrointestinal wall and adjacent structures from within the digestive tract; the pancreas is an important example of an organ examined by this route. These approaches change the acoustic access and probe geometry, while retaining the same basic pulse–echo principle. (radiologyinfo.org)

Intravascular ultrasound uses a miniature transducer carried on a catheter within a blood vessel, producing views of the vessel lumen and wall from inside rather than through overlying tissues. Intraoperative ultrasound instead places a sterile probe directly in the operative field during surgery, giving access to exposed structures. Neither route is equivalent to external scanning: intravascular acquisition requires catheter access, and intraoperative acquisition depends on surgical exposure. Probe placement determines which intervening tissues the beam must traverse and the directions from which the target can be examined. (radiologyinfo.org)

Anatomical Imaging Modes and Image Presentation

Amplitude-mode (A-mode) presents echoes along a single acoustic beam as a one-dimensional trace: position along the horizontal axis represents echo delay or its corresponding depth, while peak height represents echo amplitude. It therefore shows the locations and strengths of reflecting interfaces rather than a cross-sectional picture. Early instruments could produce this display using an analog oscilloscope. Brightness-mode (B-mode) instead encodes echo amplitude as brightness at the corresponding depth; assembling scan lines across a plane produces a two-dimensional image of anatomy. Stronger echoes generally appear brighter, whereas regions returning few or no echoes appear dark. Display brightness also depends on receiver gain and depth-dependent amplification, so it is not a direct measurement of tissue composition. (citeseerx.ist.psu.edu)

Early static B-mode scanners used articulated arms to record the position and direction of the transducer while an image was assembled. Commercial systems spread from the mid-1960s, and subsequent developments replaced bistable displays—essentially presenting echoes as either present or absent—with grayscale images and replaced static acquisition with real-time moving images. Motion-mode (M-mode) takes a different approach: it repeatedly samples one selected beam line and displays successive depth profiles across a time axis, with echo amplitude encoded as brightness. Its two axes thus represent depth and time, not two spatial dimensions. Stationary interfaces form approximately horizontal traces, while moving interfaces produce changing contours. A line through the mitral valve, for example, records valve movement over successive cardiac cycles; an accompanying B-mode image identifies where the line passes through the heart. (aium.s3.amazonaws.com)

Two-dimensional imaging depicts a section through an organ, whereas three-dimensional ultrasound combines spatially located measurements into a volume. One acquisition method tracks the position of a conventional probe as it sweeps across the target, allowing successive two-dimensional sections to be placed within a common coordinate system and reconstructed through image processing. In a 2009 study, Thomas J. MacGillivray and colleagues used optical tracking to reconstruct rectus femoris muscle volumes in 11 healthy volunteers; measured volumes ranged from 5 to 28 cubic centimetres. This illustrates that a three-dimensional dataset contains volumetric information rather than merely a perspective view of a single section. Four-dimensional (4D) imaging adds time to three-dimensional imaging, displaying successive volumes as a moving three-dimensional sequence. (pubmed.ncbi.nlm.nih.gov)

Image orientation connects the screen to the patient’s anatomy. In the American Institute of Ultrasound in Medicine’s 2020 presentation standard, the probe’s scanning surface is usually represented at the top of the image, with increasing depth below it. A notch or other probe indicator corresponds to an indicator on the display. For conventional anterior abdominal scanning, the probe indicator points toward the head in a longitudinal view and toward the patient’s right in an axial view; axial images are presented as if viewed from the feet. Labels also distinguish body planes from organ axes: a longitudinal view of the kidney, for example, need not coincide with a sagittal body plane. These conventions are not universal across all specialties or examination routes, making explicit orientation labels important. (aium.s3.amazonaws.com)

Anatomical annotations identify the organ, side, and imaging plane, while measurement annotations show what was measured. The AIUM’s fetal image library, for example, pairs left/right femur labels with measurement markers and a femur-length value in centimetres. A live display changes continuously during scanning; a frozen frame preserves one moment, whereas a saved cine loop preserves a sequence that can be replayed to assess movement. The AIUM documentation parameter published in January 2026 distinguishes still-frame records from dynamic recordings and specifies that archived images retain patient and facility identifiers, examination date and time, anatomical location, laterality, and orientation where appropriate. Digital identifiers may reside in metadata provided they remain readable during image review. (aium.org)

Doppler Assessment of Blood Flow and Tissue Motion

Moving red blood cells act as acoustic scatterers, returning echoes whose frequency differs from the transmitted frequency through the Doppler effect. For a transmitted frequency $f_0$, scatterer speed $v$, and angle $\theta$ between the beam and motion, the shift is approximately

where $c$ is conventionally taken as $1540\ \mathrm{m/s}$ in soft tissue. The factor of two reflects transmission to and reflection from the moving scatterer. Positive and negative shifts distinguish motion toward and away from the transducer. Velocity estimation therefore depends on beam alignment, not simply on the measured frequency shift. (asecho.org)

Spectral Doppler displays the distribution of Doppler shifts—or derived velocities—over time, with signal intensity represented by brightness. Pulsed-wave Doppler selects a sample volume at a particular depth; continuous-wave Doppler records high velocities without localizing their origin along the beam. Color Doppler instead overlays local estimates of mean motion and direction on a grayscale anatomical image. (asecho.org) Pulsed spectral processing commonly uses a Fourier transform, whereas color processing estimates the phase change between successive echoes by autocorrelation. These signal-processing methods produce complementary presentations: a detailed temporal waveform at a selected location versus a spatial map across a region. (asecho.org)

Pulsed measurements are subject to aliasing when the Doppler shift exceeds the Nyquist limit, equal to half the pulse repetition frequency. A spectral waveform may then wrap across the display baseline and appear to indicate reversed flow. (asecho.org) Other misleading signals arise from settings, motion, and acoustic geometry. Excessive gain can broaden a spectrum and overestimate its peak velocity. (asecho.org) An excessively high wall filter can remove clinically important low-frequency signals; unintended probe or tissue motion can create a generalized Doppler shift. Strong reflecting surfaces can duplicate apparent color flow, while side lobes can introduce signals from outside the intended sample volume. Near a $90^\circ$ interrogation angle, directional ambiguity may mimic bidirectional flow. These are distortions of motion information, distinct from grayscale artifacts affecting anatomical appearance. M. A. Pozniak, J. A. Zagzebski, and K. A. Scanlan organized these mechanisms into equipment settings, anatomical factors, and physical or technical limitations in their 1992 review. (pubmed.ncbi.nlm.nih.gov)

Carotid examination illustrates the combination of anatomical localization and quantitative Doppler sampling. Grayscale and color images identify the vessel lumen and guide placement of the spectral sample volume; waveforms provide peak systolic and end-diastolic velocities used in assessing narrowing. Vascular laboratory methods integrate these measurements with the anatomical findings rather than treating a single velocity as a complete diagnosis. (asecho.org) The 2016 American Institute of Ultrasound in Medicine practice parameter specified an interrogation angle of $45^\circ$–$60^\circ$ whenever possible and consistent angle correction between examinations. Incorrect angle assignment becomes particularly consequential above $60^\circ$, where small angular errors produce larger errors in calculated velocity. (flaus-us.org)

Tissue Doppler imaging applies the same motion-estimation principles to myocardium rather than blood. Cardiac tissue generally returns stronger echoes but moves more slowly, so tissue-mode filtering preserves signals that blood-flow processing suppresses. Spectral sampling at the mitral annulus or basal ventricular wall records tissue velocity through the cardiac cycle; color tissue Doppler maps velocities across multiple myocardial regions. Only the component parallel to the ultrasound beam is measured, not the complete three-dimensional motion. The 2011 joint American Society of Echocardiography–European Association of Echocardiography consensus specified an alignment angle no greater than $15^\circ$ for tissue velocity measurements, corresponding to less than 4% underestimation from angular misalignment. (asecho.org)

Contrast Enhancement and Tissue Stiffness Imaging

Contrast-enhanced ultrasound (CEUS) uses intravenously injected microbubbles containing gas within a stabilizing shell. Typically about 1–3 µm in diameter, the bubbles pass through capillaries but remain within the vascular compartment rather than entering the interstitial space. The gas core expands and contracts nonlinearly under acoustic pressure, producing strong echoes with harmonic components that contrast-specific imaging separates from background tissue signals. Low-output imaging limits bubble destruction, allowing continuous observation of contrast arrival, distribution, and disappearance. The resulting enhancement displays tissue perfusion, including microvascular flow that may be inconspicuous on conventional Doppler imaging. The bubbles subsequently dissolve, and their gas is eliminated through exhalation. (pubs.rsna.org)

For focal liver abnormalities, CEUS characterizes the spatial pattern of enhancement and its evolution through arterial, portal venous, and late phases. A hepatic hemangioma, for example, commonly shows peripheral nodular enhancement followed by progressive inward filling; a 2002 study documented this pattern in 15 of 20 hemangiomas. Enhancement and subsequent washout relative to the surrounding liver also contribute to the assessment of malignant lesions, although interpretation depends on lesion type and clinical context. On March 31, 2016, the US Food and Drug Administration approved sulfur hexafluoride lipid microspheres, marketed as Lumason, for characterization of focal liver lesions in adults and children. Its supporting adult studies included 499 patients and demonstrated improved lesion characterization compared with unenhanced ultrasound. (pubmed.ncbi.nlm.nih.gov)

In the kidney, CEUS helps distinguish avascular material within a complex cyst from enhancing septa or solid nodules, and can characterize abnormalities left indeterminate by other imaging. These are clinical perfusion applications, distinct from targeted molecular imaging. Targeted microbubbles carry surface ligands that bind selected molecules on vascular endothelial cells; retained bubbles therefore provide information about molecular expression rather than circulating blood alone. A first-in-human study published in 2017 investigated BR55 microbubbles targeting kinase insert domain receptor, also called VEGFR-2, in breast and ovarian lesions. Such investigational receptor-targeted imaging is separate from routine, untargeted CEUS characterization of liver and renal masses. (pubs.rsna.org)

Elastography measures or maps tissue mechanical properties rather than contrast-agent distribution. In strain elastography, external compression produces deformation, and echo tracking estimates local strain—the fractional change in tissue dimensions. Under comparable loading, stiffer regions generally deform less. Images display relative strain, sometimes supplemented by ratios between a lesion and reference tissue; because the applied stress is usually not measured locally, these are not direct measurements of absolute elastic modulus. Tsuyoshi Shiina’s 2013 Japanese Society of Ultrasonics in Medicine guidelines distinguish this approach from shear-wave methods and emphasize that both depend on assumptions about tissue properties and measurement conditions. (pubmed.ncbi.nlm.nih.gov)

Shear-wave elastography instead estimates stiffness from the speed of mechanically induced shear waves. Vibration-controlled transient elastography generates these waves with an external vibration, whereas acoustic radiation force impulse techniques use focused ultrasound pulses to displace tissue internally. Tracking pulses measure propagation speed; systems report results in metres per second or convert them into an estimated stiffness in kilopascals. Point shear-wave elastography samples a selected region, while two-dimensional techniques produce spatial stiffness maps. These distinctions matter because acquisition methods and their interpretation are not interchangeable with strain imaging. (pubs.rsna.org)

Liver stiffness measurements provide noninvasive evidence relevant to liver fibrosis and advanced chronic liver disease, including cirrhosis, but do not measure scar tissue specifically. Inflammation, obstructive cholestasis, and hepatic venous congestion can also increase stiffness. The Society of Radiologists in Ultrasound’s 2020 consensus classified acoustic-radiation-force measurements of 9–13 kPa as suggestive, rather than diagnostic, of compensated advanced chronic liver disease in specified disease settings. Stiffness also supplies supplementary information about focal breast and thyroid lesions, but overlaps between benign and malignant tissue prevent a stiffness finding alone from establishing cancer. Interpretation incorporates anatomical imaging, laboratory and clinical findings, and, where applicable, biopsy results. (pubs.rsna.org)

Obstetric, Gynecologic, and Pediatric Applications

During pregnancy, ultrasound assesses gestational age, fetal size, anatomy, and the intrauterine environment. First-trimester crown–rump length is the most accurate sonographic measurement for dating; later biometry uses biparietal diameter, head circumference, abdominal circumference, and femoral diaphysis length. These measurements feed statistical models that estimate fetal weight and permit comparison with gestational-age percentiles and previous examinations. Estimated weight is not a direct measurement: the American Institute of Ultrasound in Medicine’s (AIUM) 2024 practice parameter notes that even the best prediction methods can produce errors as large as ±15%. A standard second- or third-trimester examination also records fetal number, cardiac activity, presentation, amniotic fluid volume, placental position, cord insertion, and an anatomical survey. (doi.org)

A standard anatomical survey, a detailed anatomical examination, and fetal echocardiography have different scopes. The detailed examination includes the standard components plus additional views determined by a suspected abnormality or other clinical indication; it is not simply a longer routine scan. Fetal echocardiography systematically examines cardiac chambers, outflow tracts, venous connections, and aortic and ductal arches, adding color and pulsed-wave Doppler assessment of blood flow and M-mode or Doppler analysis of rhythm. By contrast, a limited obstetric examination answers a specific question, such as fetal presentation or cardiac activity, without constituting a comprehensive anatomical survey. AIUM’s June 16, 2020 statement explicitly distinguishes this focused examination from standard and detailed studies. (aium.org)

Gynecologic ultrasound evaluates the uterus, cervix, endometrium, ovaries, adnexal regions, and pelvic free fluid, using transabdominal and transvaginal approaches as appropriate. Applications include investigation of abnormal bleeding, pelvic pain, uterine malformations, fibroids, and adnexal masses. Endometrial thickness is measured perpendicular to the uterine longitudinal axis, excluding fluid within the cavity; ovarian dimensions are recorded in three axes, while abnormalities are characterized by their morphology and relationship to adjacent organs. In reproductive medicine, focused examinations follow follicular development during ovarian stimulation and measure endometrial thickness and appearance. AIUM’s 2019 focused-examination parameter specifies counting baseline follicles smaller than 10 mm and measuring larger follicles; later monitoring includes measurements of the three largest follicles in each ovary. Such monitoring is narrower than a comprehensive pelvic examination. (onlinelibrary.wiley.com)

Neonatal and infant neurosonography uses open fontanelles as acoustic windows into the brain. Coronal sweeps and sagittal and parasagittal images through the anterior fontanelle depict the ventricles, periventricular white matter, deep gray nuclei, and midline structures. Mastoid-fontanelle views improve visualization of the posterior fossa. Distinct applications include detecting germinal-matrix or intraventricular hemorrhage, assessing ventricular enlargement, and examining congenital malformations or suspected hypoxic–ischemic injury. Serial examinations can document changes in ventricular size or previously identified lesions. The 2024 AIUM revision notes that access may become limited after approximately 6–9 months, although the usable window depends on fontanelle size rather than age alone. (onlinelibrary.wiley.com)

Infant spinal ultrasound exploits the incomplete ossification of posterior vertebral elements, which permits visualization of the spinal canal before increasing ossification restricts access. Longitudinal and transverse imaging assesses the conus medullaris, filum terminale, cauda equina, and surrounding fluid spaces. Evaluation for spinal dysraphism or cord tethering includes identifying the vertebral level of the conus, examining cord position, and observing nerve-root motion. Accurate vertebral counting matters because a low-appearing conus cannot be interpreted without a reliable anatomical level; skin tracts may also be examined for continuity with the canal. (aium.org)

Assessment of developmental dysplasia of the hip combines static morphology with dynamic stability testing. The 2024 AIUM parameter specifies examination of both hips using a standardized coronal view and a transverse view with the hip flexed, with and without gentle stress. Coronal imaging depicts the acetabular roof and femoral-head position; the Graf alpha angle quantifies the bony roof, with a normal value of at least $60^\circ$. Dynamic imaging observes displacement and reducibility of the femoral head. A complete hip examination therefore evaluates position, acetabular morphology, and stability separately, rather than treating a single image or angle as the entire examination. (onlinelibrary.wiley.com)

Cardiovascular and Other Organ-System Applications

Echocardiography examines cardiac anatomy and function rather than primarily surveying vessels outside the heart. It depicts chamber size and shape, myocardial movement, valve function, and the pericardium. Functional measurements include ejection fraction, the proportion of ventricular blood expelled during a heartbeat. Doppler assessment helps characterize valve narrowing and backward leakage, while anatomical images can reveal abnormal communications between chambers, intracardiac clots, and growths associated with valves. These observations address different components of cardiovascular disease: impaired pumping, abnormal valve mechanics, and structural defects are not interchangeable findings. Peripheral vascular examinations instead concentrate on vessel patency, obstruction, dilatation, and blood-flow behavior in specified arterial or venous territories. (heart.org)

Arterial ultrasound combines vessel-wall and lumen imaging with flow assessment to investigate narrowing associated with atherosclerosis, occlusion, aneurysms, and the condition of bypass grafts. Abdominal aortic examinations obtain longitudinal and transverse views and record the greatest diameter; the AIUM’s 2021 parameter specifies outer-edge-to-outer-edge measurements, avoiding substitution of the residual flowing lumen for the whole vessel. Venous examinations use a different diagnostic mechanism: a normal accessible vein collapses under transducer pressure, whereas incomplete compression is an important finding in the assessment of thrombosis. The AIUM’s 2024-revision peripheral venous parameter specifies compression at intervals of 2 cm or less, supplemented by color and spectral Doppler observations. Separate reflux studies examine backward venous flow rather than merely searching for thrombus. (radiologyinfo.org)

Dialysis-access examinations address surgically created arteriovenous fistulas and synthetic grafts. Before access creation, ultrasound mapping establishes the course, size, and patency of potential vessels. Postoperative studies examine the access circuit for narrowing, thrombus, aneurysmal changes, and abnormal flow that may explain inadequate dialysis or hand ischemia. They can also quantify flow volume: the 2020 AIUM postoperative parameter identified flow below approximately 500–600 mL/min, or an interval decrease of 25%, among indications for investigating inadequate access flow. These figures describe access-assessment criteria, not universal thresholds for diagnosing vascular disease elsewhere. (onlinelibrary.wiley.com)

Abdominal examinations survey the liver, gallbladder, bile ducts, spleen, pancreas, and kidneys, with the exact scope determined by the clinical question. Examples include investigating gallstones, enlarged organs, and structural abnormalities accompanying abnormal liver tests. Urinary imaging evaluates kidney size and structure, collecting-system dilatation, stones, cysts, and masses; bladder images obtained before and after urination can assess emptying. Transplanted organs require organ-specific examinations rather than a generic abdominal survey. The AIUM’s 2024 solid-organ transplant parameter combines grayscale assessment of the graft and surrounding collections with Doppler evaluation of vascular connections and perfusion; renal graft studies additionally examine the collecting system, while liver graft studies include the biliary system. (radiologyinfo.org)

Breast ultrasound characterizes palpable or imaging-detected abnormalities, particularly whether a finding is fluid-filled, solid, or mixed. It complements mammography rather than replacing it, because some cancers present principally as mammographic calcifications. Thyroid ultrasound localizes neck lumps, depicts nodules, and measures changes in their size, but does not establish whether thyroid hormone production is normal, excessive, or deficient. Neither breast nor thyroid appearance alone invariably establishes malignancy; selected abnormalities undergo biopsy for tissue diagnosis. Scrotal imaging localizes abnormalities within or outside the testis and investigates pain, swelling, trauma, and varicoceles, with Doppler contributing information about perfusion. (radiologyinfo.org)

Musculoskeletal ultrasound depicts tendons, muscles, ligaments, nerves, joint fluid, and superficial soft-tissue masses. Real-time examination during movement can demonstrate abnormalities such as tendon displacement that may be less apparent in a stationary image. Ocular ultrasound serves distinct measurement and anatomical purposes: A-scan biometry measures axial length for intraocular-lens calculations, whereas B-scan imaging evaluates posterior ocular structures when dense cataract or vitreous hemorrhage obstructs direct visualization. In that setting, it can depict retinal detachment and intraocular masses despite loss of an optical view through the eye. (radiologyinfo.org)

Point-of-Care Imaging and Procedure Guidance

Point-of-care ultrasonography (POCUS) is defined by its use at the patient’s location to answer clinically focused questions, rather than by a separate acoustic technology. Portable equipment brings conventional ultrasound imaging to the bedside, where findings can be incorporated directly into clinical assessment. Questions may include whether the bladder is distended, whether pericardial fluid is present, or whether a proximal deep vein is compressible. The American Institute of Ultrasound in Medicine (AIUM) published a point-of-care practice parameter in 2019 covering abdominal, thoracic, cardiac, and deep-vein examinations. It characterizes POCUS as complementary to the physical examination, not as a means of detecting every abnormality; an unresolved question or an incidental finding may require a comprehensive examination. (aium.org)

Focused assessment with sonography for trauma (FAST) is a protocol-based emergency examination that searches for fluid in spaces where traumatic bleeding may accumulate. Its four principal windows assess the pericardium, the right upper abdomen—including the hepatorenal recess, or Morison’s pouch—the left upper abdomen around the spleen and left kidney, and the pelvis around the bladder. Extended FAST (eFAST) adds thoracic assessment for pleural fluid and pneumothorax. Anterior chest views examine movement at the pleural interface: lung sliding represents the relative movement of apposed pleural surfaces, whereas a “lung point” marks a transition between areas with and without sliding and supports identification of pneumothorax. AIUM lists an eFAST practice parameter dated 2022. These examinations locate selected signs of injury rather than provide a complete inventory of damaged organs; a negative FAST does not exclude injury, especially when bleeding is small-volume or confined to the retroperitoneum. (acep.org)

Procedure guidance uses ultrasound to display a target, adjacent anatomical structures, and an instrument during an intervention. In an ultrasound-guided biopsy, for example, images show the needle’s position relative to a breast mass; during vascular access, they show the instrument in relation to the vessel. Imaging can therefore provide spatial information while the procedure is taking place, rather than only identify a skin entry site beforehand. Applications also include nerve blocks associated with anesthesia. The distinction between preliminary localization and real-time guidance is important: an initial image maps the target, whereas dynamic imaging follows changes in the relationship between target and instrument as the instrument advances. (nibib.nih.gov)

Needle visualization depends on alignment with the ultrasound imaging plane. In an in-plane approach, the needle lies along that plane, allowing its shaft and tip to be displayed together when alignment is maintained. In an out-of-plane approach, the needle crosses the plane and appears as a bright cross-sectional dot. That dot can represent either the shaft or the tip, so its appearance alone does not establish the tip’s position. Needle visibility also depends on insertion angle: a steep trajectory can make the needle less conspicuous. AIUM’s selected-procedures parameter describes probe positioning and needle-tracking techniques for these geometries; the 2016 version specifies competence in both approaches. Thus, seeing part of a needle is distinct from reliably locating its advancing tip. (aium.org)

Ultrasound-guided intervention is distinct from therapeutic ultrasound. In guidance, acoustic echoes supply images while a needle, catheter, or other instrument performs the intervention. In therapeutic ultrasound, acoustic energy itself produces the intended tissue effect, such as heating, mechanical disruption, or ablation. High-intensity focused ultrasound concentrates acoustic output at a selected target; ultrasound imaging or magnetic resonance imaging may separately identify that target and monitor treatment. Consequently, a treatment can be image-guided while also using ultrasound as its therapeutic energy source, but an ordinary ultrasound-guided biopsy does not remove tissue through acoustic action. (nibib.nih.gov)

Image Artifacts, Diagnostic Limitations, and Complementary Imaging

Grayscale ultrasound images depend on assumptions about how echoes reach the transducer: sound travels along a straight path, each echo represents a single reflection, and its return time indicates the reflector’s depth. Violations can create structures that are absent, displace real structures, or obscure anatomy. Reverberation occurs when sound repeatedly bounces between reflective interfaces, including between the transducer and a tissue boundary. Successive echoes arrive progressively later and are displayed as increasingly deep, often equally spaced lines rather than repeated journeys along the same path. These lines may extend into otherwise fluid-filled spaces and mimic internal material. (pmc.ncbi.nlm.nih.gov)

Acoustic shadowing and posterior enhancement arise from differences in attenuation. A strongly reflecting or attenuating object leaves relatively little sound available to interrogate deeper tissue, producing a dark region behind it. Stones and cortical bone are common examples. Conversely, fluid-containing structures generally attenuate sound less than surrounding soft tissues; echoes from tissue behind them consequently appear brighter. Enhancement behind the gallbladder or a cyst therefore describes altered transmission, not unusually reflective tissue at that depth. These appearances can provide useful evidence about a structure’s composition: shadowing supports recognition of a stone, while enhancement supports a fluid-containing lesion. Neither appearance alone establishes a complete diagnosis or excludes other abnormalities. (doi.org)

Apparent displacement and duplication reflect errors in assigning echoes to locations. Mirror-image artifacts occur when a strong interface redirects sound along an indirect path; the scanner assigns the delayed echo to a deeper point along the assumed straight beam. A liver structure can thus appear duplicated beyond the diaphragm. Refraction bends the beam as it crosses an interface obliquely between materials with different sound speeds, potentially shifting or duplicating an apparent structure. Beam width and side lobes introduce another error: echoes originating outside the intended imaging line may be placed within it. Such echoes can simulate material within a cavity. Alternative imaging planes and changes in probe angulation help distinguish these appearances from structures with consistent anatomical relationships. (pmc.ncbi.nlm.nih.gov)

Air and bone restrict acoustic access but do not make every nearby structure inaccessible. Reflection at tissue–air boundaries limits penetration into aerated lungs and gas-filled bowel, while bone obstructs assessment of deeper soft tissues. Accessible interfaces can nevertheless yield useful echoes, and fluid-filled or partially fluid-filled regions around the lungs can be examined. The National Institute of Biomedical Imaging and Bioengineering’s April 2022 fact sheet explicitly distinguishes these situations from the difficulty of imaging normally aerated lung or bone interiors. Additional tissue depth also weakens returning signals: larger body habitus and deeply situated targets can reduce image quality. A region that is poorly visualized is not equivalent to a region demonstrated to be normal. (nibib.nih.gov)

An uncertain finding may require additional views, repeat examination, or a complementary modality. Computed tomography and magnetic resonance imaging can address questions left unresolved by an inadequate acoustic window or incomplete lesion assessment, rather than occupying a universal position above ultrasound. Conversely, contrast-enhanced ultrasound can characterize some liver abnormalities discovered on routine scanning and lesions not well seen on CT or MRI, sometimes avoiding further cross-sectional imaging. Its own limitations remain relevant: the Radiological Society of North America and American College of Radiology identify lesions smaller than 1 cm, lesions near the lung, and deep liver lesions as potentially difficult targets. Selection of complementary imaging therefore depends on the unresolved anatomical or diagnostic question and the limitations of the particular examination. (radiologyinfo.org)

Biological Effects and Safety Frameworks

Diagnostic ultrasound introduces acoustic energy into tissue without the ionizing-radiation exposure associated with X-rays and computed tomography. Its potential biological effects therefore arise from heating and mechanical interactions rather than ionization. Absorption converts acoustic energy into heat; the resulting temperature rise depends on the transmitted output, exposure duration, and properties of the insonated tissue. Oscillating pressure also produces mechanical effects, including acoustic cavitation, in which gas bubbles undergo oscillation, growth, or collapse. These mechanisms establish the need for exposure control, but their physical possibility does not mean that injury occurs during an ordinary diagnostic examination. The FDA describes ultrasound as generally safe when used prudently by appropriately trained healthcare providers. (fda.gov)

The displayed thermal index (TI) estimates the potential for heating using simplified tissue models. Numerically, it relates acoustic power to the power estimated to produce a $1\,^{\circ}\mathrm{C}$ temperature rise under specified conditions. Different models address soft tissue (TIS), bone near the beam’s focus (TIB), and cranial bone close to the transducer (TIC). A TI of 1 is consequently not a measurement showing that the patient’s tissue temperature has risen by exactly $1\,^{\circ}\mathrm{C}$. Model assumptions, tissue composition, and exposure conditions limit its correspondence with actual temperature. It is neither a direct measurement of injury nor a guarantee of safety. (aium.s3.amazonaws.com)

The mechanical index (MI) indicates the relative potential for nonthermal effects, particularly cavitation. It is calculated as $\mathrm{MI}=p_{r,0.3}/\sqrt{f_c}$, using the numerical value of derated peak rarefactional pressure in megapascals and center frequency in megahertz. Derating estimates attenuation between the transducer and tissue; the conventional coefficient is $0.3\,\mathrm{dB\,cm^{-1}\,MHz^{-1}}$. MI does not directly detect bubbles or tissue damage, and the presence of gas bodies affects how it should be interpreted. Both indices describe aspects of acoustic exposure rather than comprehensive biological risk. (bmus.org)

Professional safety frameworks apply the as low as reasonably achievable (ALARA) principle while preserving the information needed for diagnosis. Acoustic output, total scanning duration, and transducer dwell time—the period during which a particular region remains exposed—are distinct control variables. A brief examination can still concentrate exposure at one location, whereas moving the probe changes its spatial distribution. The FDA calls for minimizing exposure while maintaining diagnostic quality, and AIUM practice parameters explicitly address both dwell time and total scanning time. Monitoring TI and MI provides exposure-related feedback within this framework; neither index substitutes for controlling duration. (fda.gov)

Fetal exposure receives particular attention. In its May 19, 2020 statement, the American Institute of Ultrasound in Medicine (AIUM) reported insufficient evidence to establish a causal relationship between diagnostic ultrasound and recognized adverse human effects, while emphasizing that absence of demonstrated harm is not proof that adverse effects are impossible. For documenting fetal cardiac activity or heart rate, the statement favors B-mode or M-mode, with TI preferably no greater than 0.7, rather than spectral Doppler unless clinically indicated. AIUM discourages fetal scanning without a medical indication. The FDA distinguishes commercial “keepsake” examinations from images supplied during a medically indicated examination without additional exposure. (aium.org)

Contrast administration and invasive procedures introduce separate risks. Injected microbubbles provide gas bodies that can participate in cavitation; AIUM’s August 19, 2021 statement describes experimentally observed microvascular and cellular effects during contrast exposure at MI values above 0.4, while noting uncertainty about their medical significance. Some agents contain polyethylene glycol, which can rarely cause allergic reactions. These drug-related and bubble-mediated risks differ from the bleeding, infection, and accidental injury to adjacent structures associated with ultrasound-guided biopsy. The use of nonionizing imaging guidance does not eliminate complications of the needle procedure itself. (aium.org)

Examination Quality, Documentation, and Professional Practice

Acquisition and interpretation are distinct professional responsibilities, although the same clinician may perform both. The American Institute of Ultrasound in Medicine (AIUM) requires interpreting providers in accredited practices to have training and experience in the specific area of sonography, supported by documented continuing medical education and examination volumes. Its accreditation categories distinguish, for example, standard obstetric examinations, detailed fetal anatomical examinations, fetal echocardiography, and ultrasound-guided regional anesthesia. Training guidelines therefore define competence within an examination’s scope rather than treating proficiency in one application as sufficient for every application. A designated ultrasound director oversees the quality and appropriateness of the practice’s operations. (aium.org)

Practice parameters specify the structures, views, and documentation that constitute a complete examination in a particular area. AIUM explicitly describes these as minimum examination criteria, not legally defined standards of care; deviations may be necessary because of patient needs or available equipment. (aium.org) The AIUM documentation parameter, revised in 2025 and published on January 10, 2026, requires a permanent, retrievable record of the examination and its interpretation. Relevant normal and abnormal areas are recorded, using still images, cine loops, or both as appropriate. Archived images or their displayable metadata identify the patient, facility, examination date and time, anatomical location, laterality, and orientation where applicable, together with the thermal and mechanical indices. These records permit later assessment of diagnostic quality rather than merely preserving selected abnormal findings. (onlinelibrary.wiley.com)

The signed final report identifies the examination performed, its clinical indication, and any endocavitary approach. It documents findings, appropriate anatomical measurements, measurements of abnormal structures when taken, and an impression or differential diagnosis where appropriate. Contrast agents and medications are recorded with route and dose when applicable, along with significant reactions or complications. Omitted standard views or other required components must be accounted for, and relevant prior imaging is compared when available. When delayed communication could adversely affect an outcome, or preliminary and final interpretations differ significantly, the record includes the date, time, and method of provider-to-provider communication. Images and reports are retained according to clinical needs and applicable legal and institutional requirements. (onlinelibrary.wiley.com)

Equipment quality assurance examines both physical integrity and imaging performance. AIUM’s accreditation manual calls for documented testing at least annually, more frequently when problems arise, and quality control of repaired transducers. (aium.org) Its equipment-testing framework includes inspection of probe housings and cables, display performance, image uniformity, sensitivity, and geometric measurement accuracy. A tissue-mimicking test object or transducer tester can reveal nonuniformity associated with inactive array elements; known distances provide a check on measurement calibration. (aium.s3.amazonaws.com) Personnel monitoring is complementary: AIUM’s April 20, 2026 accreditation standards require regular retrospective performance reviews of interpreting providers and sonographers, with written documentation. (aium.org)

Transducer preparation depends on the tissues contacted and the procedure’s sterility requirements. AIUM’s July 2, 2025 revision specifies cleaning followed by low-level disinfection for external probes used on intact skin, and high-level disinfection plus a single-use cover for internal examinations. A cover does not replace high-level disinfection because barrier integrity cannot be assumed. Cleaning removes gel and debris before disinfection; device compatibility, contact conditions, rinsing, drying, and storage follow manufacturer instructions. For percutaneous procedures, covers must meet the procedure’s sterility requirements—for example, a sterile cover for biopsy—and sterile gel is used. (aium.org) Organizational patient-safety policies also address patient identification, an invasive-procedure time-out, verification of the correct procedure and location, specimen labeling and transfer, infection prevention, confidentiality, and reporting of incidents or complications. (aium.org)

Historical Development and Research Directions

In 1942, the Austrian physician Karl Theodore Dussik published work on transmitting ultrasound through the head to investigate the brain, an early medical application distinct from the pulse-echo imaging subsequently adopted in clinical scanners. His approach examined transmitted sound rather than constructing images from returning echoes. Medical ultrasound developed through work in several countries: researchers in the United States, Japan, and Europe pursued different instruments and applications during and after the Second World War. The Glasgow group led by Ian Donald became particularly influential in turning these experimental approaches into practical diagnostic equipment during the mid-1950s. (bmus.org)

On June 7, 1958, Ian Donald, John MacVicar, and Tom Brown published “Investigation of Abdominal Masses by Pulsed Ultrasound” in The Lancet, pages 1188–1195. The paper documented the investigation of abdominal masses using pulsed sound, establishing a practical connection between ultrasonic instrumentation and clinical diagnosis. Its historical importance lay in demonstrating an approach that could be developed into usable examinations, rather than in introducing every component of ultrasound technology. The Glasgow work helped accelerate subsequent adoption, particularly in obstetrics and gynecology; the British Medical Ultrasound Society commemorates the collaboration through its Donald MacVicar Brown Lecture. (pubmed.ncbi.nlm.nih.gov)

Commercial systems became more widely available from the mid-1960s. Improvements in electronics and piezoelectric materials supported the transition from bistable displays, which represented echoes with two brightness states, to grayscale images capable of displaying differences in echo strength. Another transition replaced slowly assembled still images with continuously refreshed, real-time displays. Doppler techniques developed alongside anatomical imaging; duplex scanning combined these capabilities, and color Doppler subsequently incorporated flow information into anatomical images. From the 1970s, microchips and increasing processing capacity supported digital beamforming, signal enhancement, and additional displays, including power Doppler and three-dimensional imaging. (bmus.org)

Research extends both the hardware and the computational stages of imaging. The US National Institute of Biomedical Imaging and Bioengineering identifies transducer materials and arrays, acquisition methods, signal processing, functional imaging, and image fusion as research areas, alongside machine learning for acquisition and interpretation. A concrete hardware example appeared in 2023, when Hongjie Hu and colleagues reported a wearable cardiac ultrasound imager. Its conformable construction improved coupling to the skin and enabled imaging of the left ventricle during movement. A deep-learning model extracted ventricular volume from image sequences to derive stroke volume, cardiac output, and ejection fraction. This was a research demonstration of continuous wearable imaging, not evidence that such patches had replaced conventional clinical echocardiography. (nibib.nih.gov)

Functional ultrasound imaging investigates physiological changes rather than anatomy alone. In 2011, Emilie Macé, Gabriel Montaldo, Mickael Tanter, and colleagues demonstrated high-frame-rate plane-wave imaging of transient cerebral blood-volume changes in rats, including responses to whisker stimulation and the propagation of epileptiform activity. A subsequent methodological study described acquisition at approximately 1,000 frames per second and vascular maps obtained in as little as 320 milliseconds per slice. Combining echoes from multiple transmissions and averaging many rapidly acquired images increased sensitivity to small-vessel blood signals. These experiments measured hemodynamic correlates of activity, not electrical firing directly, and the methodological study used rats with cranial openings. (nature.com)

Image fusion has also progressed beyond laboratory demonstrations in selected applications. It spatially aligns ultrasound with another imaging dataset, allowing information from magnetic resonance imaging to guide a procedure performed under live ultrasound. In a 2015 NIH study, investigators analyzed 893 MRI–ultrasound fusion-guided prostate biopsy encounters in 791 patients examined between 2007 and 2013. The dataset comprised 4,688 biopsy cores from 2,344 targets. Such clinical studies distinguish an implemented fusion workflow from experimental imaging concepts, while research programs continue to investigate improved acquisition, processing, and integration of imaging information. (pubmed.ncbi.nlm.nih.gov)

References

1. Ultrasoundnibib.nih.gov
2. Diagnostic Radiology Physics: A Handbook for Teachers and Studentspub.iaea.org
3. TG-128: Quality Assurance for Prostate Brachytherapy Ultrasoundaapm.org
4. AAPM Medical Physics Curriculumaapm.org
5. Quick, pain-free physics review: Part 2, for radiation oncology physicistsamos3.aapm.org
6. AAPM Medical Physics Curriculumaapm.org
7. Ultrasound Signal Processing: From Models to Deep Learningarxiv.org
8. Ultrasound: Diagnostic and Interventionalnibib.nih.gov
9. Ultrasound Imagingfda.gov
10. Signal Processing Overview of Ultrasound Systems for Medical Imaging-White Paperti.com
11. Design and Implementation of Analog-Digital Hybrid Beamformers for Low-Complexity Ultrasound Systems: A Feasibility Studypmc.ncbi.nlm.nih.gov
12. Pelvis Ultrasoundradiologyinfo.org
13. Appropriate use of GI endoscopyasge.org
14. Endoscopic Ultrasound and Related Technologies for the Diagnosis and Treatment of Pancreatic Disease: Research Gaps and Opportunitiesniddk.nih.gov
15. Intravascular Ultrasoundradiologyinfo.org
16. Ultrasound Physics and Technical Facts for the Beginneracep.org
17. The History of Ultrasoundbmus.org
18. Standard Presentation and Labeling of Ultrasound Imagesaium.s3.amazonaws.com
19. 3D freehand ultrasound for in vivo determination of human skeletal muscle volumepubmed.ncbi.nlm.nih.gov
20. Ultrasoundnibib.nih.gov
21. Detailed Anatomy Imagesaium.org
22. Recommendations for Quantification of Doppler Echocardiographyasecho.org
23. Current and Evolving Echocardiographic Techniques for the Quantitative Evaluation of Cardiac Mechanics: ASE/EAE Consensus Statement on Methodology and Indicationsasecho.org
24. Spectral and color Doppler artifacts.pubmed.ncbi.nlm.nih.gov
25. Noninvasive Vascular Lab Testingasecho.org
26. Contrast Materialsradiologyinfo.org
27. Enhancement of hepatic hemangiomas with levovist on coded harmonic angiographic ultrasonographypubmed.ncbi.nlm.nih.gov
28. New Pediatric Labeling Information Database - Detailaccessdata.fda.gov
29. Lumason prescribing information, 2016accessdata.fda.gov
30. Ultrasound Molecular Imaging With BR55 in Patients With Breast and Ovarian Lesions: First-in-Human Resultspmc.ncbi.nlm.nih.gov
31. JSUM ultrasound elastography practice guidelines: basics and terminologypubmed.ncbi.nlm.nih.gov
32. JSUM ultrasound elastography practice guidelines: liverpubmed.ncbi.nlm.nih.gov
33. AIUM Practice Parameter for the Performance of Detailed Second- and Third-Trimester Diagnostic Obstetric Ultrasound Examinationsaium.org
34. Limited Obstetrical Ultrasound Examinationaium.org
35. AIUM Practice Parameter for the Performance of an Ultrasound Examination of the Female Pelvisaium.org
36. AIUM Practice Parameter for the Performance of a Focused Ultrasound Examination in Reproductive Endocrinology and Female Infertilityaium.org
37. AIUM Practice Parameter for the Performance of an Ultrasound Examination of the Neonatal and Infant Spineaium.org
38. Vascular Ultrasoundradiologyinfo.org
39. AIUM Practice Parameter for the Performance of Diagnostic and Screening Ultrasound Examinations of the Abdominal Aorta in Adultsaium.org
40. Abdominal Ultrasound Examradiologyinfo.org
41. Renal (Kidney) Ultrasoundradiologyinfo.org
42. Breast Ultrasoundradiologyinfo.org
43. Thyroid Ultrasoundradiologyinfo.org
44. Ultrasound Imaging of the Scrotumradiologyinfo.org
45. Point of Care Ultrasound - AIUMaium.org
46. AIUM Practice Parameter for the Performance of Point-of-Care Ultrasound Examinationsaium.org
47. FAST | Sonoguideacep.org
48. EFASTaium.org
49. AIUM Practice Parameter for the Performance of Selected Ultrasound-Guided Proceduresaium.org
50. Fact or Artifact in Two-Dimensional Echocardiography: Avoiding Misdiagnosis and Missed Diagnosispmc.ncbi.nlm.nih.gov
51. Clinical ultrasound physicspmc.ncbi.nlm.nih.gov
52. Abdominal Contrast Enhanced Ultrasound (CEUS)radiologyinfo.org
53. Recommendedaium.s3.amazonaws.com
54. BMUS Guidelines for the Safe Use of Diagnostic Ultrasound Equipmentbmus.org
55. Section 7—Discussion of the Mechanical Index and Other Exposure Parameterspmc.ncbi.nlm.nih.gov
56. AIUM Practice Parameter for the Performance of Detailed Diagnostic Obstetric Ultrasound Examinations Between 12 Weeks 0 Days and 13 Weeks 6 Daysaium.org
57. Prudent Use and Safety of Diagnostic Ultrasound in Pregnancyaium.org
58. Statement on Biological Effects in Tissues with Ultrasound Contrast Agentsaium.org
59. Biopsy Overviewradiologyinfo.org
60. Standards and Guidelines for the Accreditation of Ultrasound Practicesaium.org
61. Practice Parametersaium.org
62. AIUM Accreditation Application Manualaium.org
63. AIUM Routine Quality Assurance of Clinical Ultrasound Equipment: Version 2.0aium.s3.amazonaws.com
64. Guidelines for Cleaning and Preparing External- and Internal-Use Ultrasound Transducers and Equipment Between Patients as Well as Safe Handling and Use of Ultrasound Coupling Gel, 2025 Revisionaium.org
65. The Donald MacVicar Brown Lecturebmus.org
66. Investigation of abdominal masses by pulsed ultrasoundpubmed.ncbi.nlm.nih.gov
67. A wearable cardiac ultrasound imagernature.com
68. Functional ultrasound imaging of the brainnature.com
69. Functional ultrasound imaging of the brain: theory and basic principlespubmed.ncbi.nlm.nih.gov
70. Comparison of magnetic resonance imaging and ultrasound (MRI-US) fusion-guided prostate biopsies obtained from axial and sagittal approachespubmed.ncbi.nlm.nih.gov