How Does Ultrasound Imaging Work?

In 1880, Pierre Curie and his brother Jacques discovered something remarkable: certain crystals, when squeezed, produced a voltage. Squeeze them again with the opposite charge and they pulsed. They had found a two-way street between electricity and mechanical motion. Pierre’s wife Marie Curie reportedly suggested the effect might one day be useful for imaging the inside of the human body. She was right, but it took nearly a century of engineering before the first clinical ultrasound images appeared in the 1950s.

Today, ultrasound is the most frequently used imaging modality in medicine worldwide. Emergency physicians use it to detect internal bleeding in seconds. Cardiologists watch heart valves open and close in real time. Surgeons guide needles to tumors with it. And millions of expectant parents hear a heartbeat for the first time through it. Understanding how it works means understanding a chain of physics tricks: converting electricity to sound, sound to echoes, and echoes to an image your doctor can read in real time.

The short answer

Ultrasound imaging works by emitting high-frequency sound waves into the body and measuring how long and how strongly they return after bouncing off tissue boundaries. A probe containing piezoelectric crystals converts electrical pulses into mechanical vibrations, then converts the returning echoes back into electrical signals. A computer calculates the depth and density of each tissue layer from the echo timing and amplitude, building a grayscale cross-sectional image in fractions of a second.

The full picture

The piezoelectric effect: the heart of the machine

The word “piezo” comes from the Greek for “push.” Piezoelectric materials, typically ceramics like barium titanate or synthetic crystals like lithium niobate, deform slightly when an electric current passes through them. Apply a voltage and they bend. Remove the voltage and they spring back. Reverse the voltage and they compress.

In an ultrasound probe, the electronics send rapid electrical pulses to these crystals at frequencies between 1 and 20 megahertz. The crystals respond by vibrating and producing a burst of sound waves. The same crystals then switch roles, absorbing incoming echoes and generating a tiny electrical signal in response. This two-way conversion happens thousands of times per second as the probe sweeps across the body.

The choice of frequency is a engineering tradeoff. Higher frequencies (closer to 20 MHz) produce sharper images with excellent resolution but penetrate only a few centimeters into the body. Lower frequencies (1 to 5 MHz) travel deeper but produce coarser images. A cardiac probe might use 2.5 MHz to see through the chest wall. A vascular probe at 10 MHz sacrifices depth for the resolution needed to track fine blood vessel walls.

Sound navigating tissue

Sound waves travel through the body at roughly 1,540 meters per second in soft tissue, though this varies slightly depending on density and compressibility. When a wave encounters a boundary between two tissues with different acoustic impedance, some of the wave reflects and some continues deeper.

Acoustic impedance is a product of tissue density and the speed of sound through it. The greater the difference between two adjacent tissues, the stronger the reflected echo. Fluid-filled structures appear dark because very little sound reflects at the fluid-tissue boundary. Bone appears white-hot because the impedance difference between soft tissue and bone is extreme, reflecting most of the sound and leaving shadowing behind it. This is why ultrasound cannot see through bone or through the lungs, which are filled with air.

The probe emits a pulse of sound and then listens. The listening window is precisely timed: a sound wave traveling at 1,540 m/s takes about 13 microseconds to reach a structure 1 centimeter deep and return. The machine fires thousands of pulses per second, building a continuous image as it sweeps. This is why ultrasound is real-time in a way that CT and MRI are not.

Anatomy of the probe

A typical ultrasound probe is a handheld wand containing an array of up to 256 piezoelectric elements arranged in a row. The electronics do not fire all elements simultaneously. Instead, they fire groups of elements in rapid sequence, creating a sweeping beam that covers a fan-shaped slice of tissue. This is called phased array scanning.

The same elements receive the echoes. A beamformer in the machine calculates the precise timing differences between echoes arriving at different elements, then sums them to reconstruct a single scan line. Thousands of these scan lines, fired and received in microseconds, compose the complete cross-sectional image you see on screen.

The probe also contains a layer of acoustic coupling gel. This substance has acoustic impedance similar to skin, which prevents the vast majority of the sound from reflecting at the air-skin boundary. Without gel, 99.9% of the ultrasound energy would reflect immediately, leaving nothing to image the body’s interior.

Beyond the grayscale: the different modes

The most common mode, called B-mode (brightness mode), produces the familiar grayscale cross-sections. But the same basic physics supports several other imaging approaches.

M-mode (motion mode) freezes the B-mode image and displays only the motion of structures over time, at high frame rates. Cardiologists use M-mode to measure the precise dimensions of heart chambers and track how quickly they dilate or contract.

Doppler ultrasound measures the frequency shift of echoes returning from moving blood. As red blood cells approach the probe, the returning echoes compress slightly, increasing in frequency. As they move away, the frequency decreases. The machine calculates this shift and displays blood flow as color overlaid on the grayscale image: red typically indicates flow toward the probe, blue indicates flow away. This is how technicians measure how fast blood is moving through a vessel and in which direction, critical for detecting blockages, valve leaks, and shunts.

3D and 4D ultrasound use electronic sweeping to acquire hundreds of parallel 2D slices and reconstruct a volume. 4D adds the dimension of time, producing a live 3D image. Obstetricians use it to visualize fetal anatomy in detail that 2D imaging cannot match. Interventional radiologists use it to guide procedures in three dimensions.

Elastography: feeling stiffness remotely

A newer technique called elastography maps the stiffness of tissue rather than just its acoustic properties. The probe applies a slight compression or generates an internal vibration, then measures how quickly tissue deforms and returns to shape. Stiff tissue (often a sign of scarring or tumor) deforms less than soft tissue. Elastography is particularly useful for characterizing liver fibrosis without a biopsy.

Why it matters

Ultrasound is unique among major imaging technologies in three respects. It produces images in real time, at the bedside, without radiation, and at a fraction of the cost of MRI or CT. A portable ultrasound machine weighs under 5 kilograms and can fit in a backpack. This is why air ambulances carry them, why emergency departments have them mounted on walls, and why some primary care physicians now own pocket-sized probes that plug into smartphones.

The global ultrasound market generates over $8 billion annually, but its greatest impact may be in low-resource settings where CT scanners and MRI machines are simply unavailable. A 2022 study published in The Lancet found that point-of-care ultrasound in rural sub-Saharan Africa reduced maternal mortality from postpartum hemorrhage by 24%, simply by enabling faster diagnosis. In the hands of a trained midwife, a $2,000 probe can identify internal bleeding that would otherwise require a day of travel to a hospital with a CT scanner.

For expectant parents, the stakes are more personal. The first ultrasound typically happens between 8 and 14 weeks of pregnancy, measuring the fetal heartbeat and confirming due dates. The mid-pregnancy anatomy scan at 18 to 22 weeks examines every major organ system, checks for structural abnormalities, and, increasingly, reveals the baby’s sex. Many parents describe hearing the heartbeat for the first time as the moment pregnancy becomes real. Behind that moment is a century of physics, a probe full of vibrating crystals, and a computer turning echoes into an image.

Common misconceptions

“Ultrasound is completely risk-free with no limits.” Ultrasound is safer than ionizing imaging, but it is not without biological effects. At high output levels, ultrasound can cause slight tissue warming and, in extreme cases, cavitation (tiny bubble formation in tissue fluid). The FDA has set limits on the mechanical and thermal output of fetal imaging devices specifically because prolonged exposure at maximum power is not well studied. Professional societies recommend using the lowest output necessary and limiting scanning time. The ALARA principle (As Low As Reasonably Achievable) governs all medical ultrasound use.

“If an ultrasound looks normal, everything is fine.” Ultrasound has real blind spots. Bone blocks sound entirely, which is why the brain cannot be imaged through the adult skull. Air-filled organs like lungs and bowel gas reflect sound so strongly that abnormalities behind them are invisible. A normal ultrasound does not rule out disease in regions the probe cannot see. Follow-up imaging with CT or MRI is common when ultrasound raises a concern in a structure it cannot fully visualize.

“Doppler ultrasound measures blood flow directly.” Doppler measures the frequency shift of echoes returning from moving red blood cells, not the flow itself. The calculation assumes a consistent angle between the sound beam and the direction of blood flow. When that angle exceeds 60 degrees, the velocity estimate becomes increasingly inaccurate. This is why sonographers angle the probe carefully: getting the angle right is what separates a reliable measurement from a misleading one.

Key terms

Acoustic impedance: A measure of how resistant a tissue is to sound wave propagation, determined by its density and the speed of sound through it. Greater differences in acoustic impedance between two adjacent tissues produce stronger echoes.

Piezoelectric effect: The property of certain crystals to generate an electrical voltage when mechanically deformed, and to deform when an electrical voltage is applied. This two-way conversion is the foundation of all ultrasound imaging.

Echo: A sound wave that has reflected off a tissue boundary and returned to the probe. The time delay between the emitted pulse and the received echo encodes the depth of the reflecting surface.

Doppler shift: The change in frequency of an ultrasound wave reflected off a moving object. The shift is positive when the object moves toward the probe and negative when it moves away, allowing blood flow velocity to be calculated.

Phased array: An ultrasound probe design in which many piezoelectric elements are fired in precisely timed sequences to steer and focus the ultrasound beam electronically, without moving parts.

ALARA principle: “As Low As Reasonably Achievable” — the guiding standard for minimizing ultrasound exposure to the lowest level necessary for diagnostic purposes.