Ultrasound Invention: How Medical Ultrasound Was Developed

Medical ultrasound was not invented on one date by one person. The technology emerged when several previously separate ideas were made to work together: high-frequency sound had to be generated and detected, echoes from biological tissue had to be measured reliably, those signals had to be converted into useful spatial information, and the resulting equipment had to become practical enough for routine examination of patients.

This distinction matters because different pioneers solved different parts of the problem. Karl Dussik made some of the earliest attempts to use ultrasound diagnostically in the human body. George Ludwig investigated how ultrasonic pulses behaved in tissue. John Wild and John Reid worked on tissue characterization and two-dimensional echo imaging. Douglass Howry produced cross-sectional images with immersion scanners. Inge Edler and Hellmuth Hertz recorded moving structures in the heart. Ian Donald, John MacVicar and Tom Brown helped turn compound contact scanning into a workable clinical method in obstetrics and gynaecology. Shigeo Satomura opened another branch by applying the Doppler effect to medical ultrasound.

The familiar modern scanner grew from the combination and refinement of these lines of work rather than from a single prototype.

Why Medical Ultrasound Had to Be Developed in Stages

Sending sound into the body is only the first part of an ultrasound examination. A useful diagnostic system must perform several different tasks in sequence.

First, the instrument needs a source capable of producing sound at frequencies above ordinary human hearing. It must then transfer that energy efficiently into tissue. When the sound encounters boundaries between materials with different acoustic properties, part of its energy returns toward the transducer. The instrument has to detect these weak echoes, measure when they arrive and determine their relative strength.

Even that does not automatically create an image. A single echo may indicate that a reflecting boundary exists at a certain depth, but it does not show the shape of an organ. Producing anatomical images requires repeated measurements from known positions and directions. Display technology must then convert those measurements into a spatial representation that a clinician can interpret.

Motion creates another problem. A static image cannot fully describe a beating heart or moving fetus. Blood flow requires yet another type of measurement, using frequency changes in reflected sound rather than simply plotting structural echoes.

For this reason, asking who invented ultrasound can refer to several different achievements: the first diagnostic experiment, the first useful pulse-echo measurements in tissue, the first two-dimensional image, the first cardiac recording, the first practical obstetric scanner or the first Doppler system.

Piezoelectricity Made an Ultrasound Transducer Possible

The physical basis of the ultrasound probe predates diagnostic imaging by decades. In 1880, Jacques and Pierre Curie demonstrated the piezoelectric effect in certain crystals. Mechanical stress could produce an electrical charge. The converse behavior, in which an electrical field causes a piezoelectric material to deform, provided a way to turn electrical signals into mechanical vibrations.

That two-way conversion is well suited to ultrasound. A transducer can use an electrical pulse to create a brief high-frequency vibration. After transmitting sound into a material or the body, the same general class of material can convert returning mechanical vibrations into electrical signals that electronics can amplify and analyze.

The early connection between piezoelectric materials and underwater acoustics became especially clear during the development of echo-ranging systems in the early twentieth century. Paul Langevin and other researchers explored quartz-based ultrasonic devices for detecting objects underwater. These systems were not medical scanners, but they showed that high-frequency sound could reveal objects that could not be seen directly.

Medical ultrasound therefore did not begin with the invention of a new kind of wave. It began when existing acoustic and electronic methods were adapted to the much harder problem of extracting useful information from living tissue.

Industrial Flaw Detection Provided a Direct Technical Predecessor

A ship plate, pressure vessel or metal component may look intact while containing an internal crack. Engineers needed ways to inspect such materials without cutting them open. Ultrasonic testing offered an answer: transmit a pulse into the material and look for echoes from hidden interfaces or defects.

Floyd Firestone’s ultrasonic reflectoscope, developed during the 1940s, belonged to this industrial line of work. Its purpose was the nondestructive examination of opaque solids rather than medical diagnosis. The idea behind it, however, was closely related to later pulse-echo medicine. A pulse entered a material, an internal boundary returned part of the energy, and the timing of that return contained information about where the reflector was located.

This industrial connection appears repeatedly in ultrasound history. Several early medical investigators either modified industrial flaw detectors or began experiments using machines built for metal inspection.

Radar is sometimes presented as though it simply became medical ultrasound. That description is technically misleading. Radar works with electromagnetic radiation, whereas ultrasound uses mechanical waves that propagate through matter. Wartime radar research contributed electronic concepts and display techniques, but ultrasonic nondestructive testing and sonar provide the closer acoustic parallels.

Karl Dussik Tried to Image the Brain With Transmitted Ultrasound

Austrian neurologist Karl Theodore Dussik is associated with some of the earliest documented attempts to use ultrasound as a diagnostic imaging method. His work on the brain began before the mature pulse-echo systems that later dominated medical imaging.

Dussik’s approach relied largely on through-transmission. Instead of sending a pulse into the body and listening for an echo returning to the same side, ultrasound passed through the head from one transducer toward a receiver positioned on the opposite side. Changes in transmitted energy were recorded and used to create what Dussik interpreted as images related to intracranial structures.

In the 1940s he described this approach as hyperphonography. A 1947 publication on brain hyperphonography included Karl Dussik, F. Dussik and L. Wyt as authors. The experimental arrangements placed the head in a water-coupled setup so that sound could travel between the transducers and the skull without a large air gap.

The work was ambitious, but the resulting images did not provide the reliable view of brain anatomy that Dussik hoped for. Later experiments showed that much of the recorded variation could be explained by reflection and attenuation caused by the skull itself. Bone presents a difficult acoustic barrier, and changes in skull thickness and geometry can strongly alter transmitted ultrasound.

That limitation changes how Dussik’s achievement should be described. Calling his work the first successful modern brain scan would overstate what the method could actually resolve. It is better understood as one of the earliest serious attempts to transform ultrasound measurements into diagnostic images of a living internal organ.

George Ludwig Established How Echoes Behave in Tissue

George D. Ludwig’s research in the late 1940s moved ultrasound closer to the pulse-echo logic used in later diagnostic systems. Working at the U.S. Naval Medical Research Institute, Ludwig investigated the acoustic characteristics of biological tissues and the possibility of detecting objects such as gallstones.

His equipment originated from ultrasonic methods used for nondestructive testing. In experimental work, Ludwig used A-mode displays, in which returning echoes appeared as peaks rather than as anatomical pictures. This was enough to investigate how strongly different tissues and objects reflected ultrasound and how the timing of an echo related to its location.

Ludwig and Francis Struthers reported work on the detection of gallstones and foreign bodies in tissue in 1949. Ludwig’s measurements also helped establish an average sound velocity in mammalian soft tissue of about 1,540 meters per second, a value that became the conventional assumption used by diagnostic scanners when converting echo travel time into distance.

The measurement illustrates an important feature of ultrasound imaging. A scanner does not directly see depth. It measures how long sound takes to make a round trip between the transducer and a reflecting interface. To translate that interval into a location on the display, the machine assumes a propagation speed for soft tissue.

Ludwig’s experiments were therefore different from Dussik’s transmission images. They helped establish the quantitative pulse-echo behavior needed to turn reflected sound into depth information.

A-Mode, B-Mode and M-Mode Solved Different Display Problems

The terms A-mode, B-mode and M-mode describe different ways of representing ultrasonic echoes. They arose because early researchers were trying to extract different kinds of information from the same physical phenomenon.

A-mode could reveal that reflecting structures were present at particular depths, but a line of spikes did not resemble an organ. B-mode added spatial organization. Echoes could be represented as luminous points, with stronger returns appearing brighter. By recording many such lines from different positions or directions, researchers could assemble a two-dimensional cross-section.

M-mode addressed motion. Repeating measurements along the same ultrasound line and plotting them over time made moving structures produce characteristic traces. This technique became especially valuable for the heart, where static anatomy alone was not enough.

John Wild and John Reid Used Echoes to Distinguish Tissue

Surgeon John Julian Wild approached ultrasound from another direction. He wanted to know whether reflected sound could reveal differences in biological tissue, including differences between normal tissue and tumors.

Wild initially used A-mode methods to measure structures such as the intestinal wall and investigate abnormal tissue. He then worked with electrical engineer John M. Reid on equipment capable of producing two-dimensional images.

In 1952, Wild and Reid published work applying echo-ranging methods to biological tissue and produced two-dimensional B-mode images in their experiments. Their research on breast and other soft-tissue lesions helped establish the idea that ultrasound might reveal more than the location of one isolated boundary. Patterns of echoes could carry information about tissue structure.

Their work also exposed a tradeoff that remains part of ultrasound physics. Higher frequencies can provide finer spatial detail, but they are attenuated more rapidly in tissue and therefore do not penetrate as deeply. Lower frequencies generally reach farther but sacrifice some spatial resolution. Modern systems manage this tradeoff with improved transducers, electronics and signal processing, but the underlying physical relationship remains.

Douglass Howry Turned Echoes Into Cross-Sectional Anatomy

At roughly the same period, radiologist Douglass Howry concentrated on the problem of producing recognizable cross-sections of the human body. His systems used reflected ultrasound, but the geometry of the scan was very different from the compact hand-held examinations familiar today.

Howry and his collaborators developed immersion systems in which the patient, body region or specimen was acoustically coupled through water. In the motorized Somascope, a transducer could move around the subject and collect echoes from different directions. Combining multiple scan lines reduced some unwanted echoes and allowed the construction of cross-sectional images known as somagrams.

Howry’s group published early diagnostic-quality two-dimensional cross-sectional images in the early 1950s. The method demonstrated that pulse-echo ultrasound could depict recognizable internal anatomy rather than merely generate a sequence of spikes.

The water bath solved one technical problem while creating several practical ones. Air is a poor medium for transferring ultrasound from a probe into the body because the acoustic mismatch at an air–tissue boundary causes strong reflection. Water provided effective acoustic coupling. Yet requiring patients or body regions to be positioned in an immersion apparatus made scanning slow, awkward and difficult to fit into ordinary clinical practice.

Howry’s later pan-scanner reduced the need for full-body immersion by placing a water-filled scanning assembly against the patient, but the larger lesson was already clear: two-dimensional ultrasound worked, although a more convenient way of coupling and moving the transducer was needed.

Edler and Hertz Made the Beating Heart Visible as Motion

The development of echocardiography followed a separate clinical problem. Swedish cardiologist Inge Edler wanted better information about diseased mitral valves before surgery. Cardiac catheterization could provide useful data, but it was invasive. Edler began exploring whether reflected ultrasound could reveal cardiac structures through the chest wall.

He collaborated with physicist Hellmuth Hertz at Lund University. In 1953 they experimented with an industrial ultrasonic reflectoscope borrowed for the work. Echoes from inside the heart could be seen on the instrument’s display.

The important step was not simply detecting an echo from the chest. The heart moves continuously. By repeatedly recording echoes from the same line and displaying position over time, Edler and Hertz could create traces corresponding to moving cardiac structures.

The first M-mode recording associated with their work dates to 29 October 1953, and their early results were published in 1954. Investigation of the mitral valve became one of the first clinical applications. Later experiments helped identify which anatomical structures produced particular moving traces.

Their equipment again shows how industrial instrumentation entered medicine. A question that had originally been framed as “Is there a flaw inside this material?” became “Where is this moving structure inside the chest, and how does its position change with time?”

Ian Donald, Tom Brown and John MacVicar Made Contact Scanning Clinically Practical

Ian Donald is often described as the inventor of medical ultrasound, especially in accounts centered on pregnancy. That wording leaves out both earlier diagnostic research and the engineering collaboration that made his own work possible.

Donald became Regius Professor of Midwifery at the University of Glasgow in 1954. He was interested in whether reflected ultrasound could help distinguish abdominal and pelvic masses. Like other pioneers, he initially experimented with technology derived from industrial flaw detection.

The clinical idea required new engineering. Tom Brown, a young engineer working for Kelvin & Hughes, became involved in 1956 after learning about Donald’s experiments. Brown understood the industrial ultrasound equipment and began designing a scanning system suitable for medical use.

The resulting work moved away from immersion-tank scanning. Brown built an early direct-contact B-mode scanner in 1956. The transducer could be moved over the patient while its position was mechanically related to the location of the echoes shown on the display. An acoustic coupling medium between the transducer and skin removed the air interface that would otherwise reflect much of the transmitted energy.

This seemingly simple change had major practical consequences. The patient no longer had to be placed in a large tank of water. The examiner could scan the body surface directly and build a compound image from multiple directions.

John MacVicar, an obstetrician working with Donald, was part of the clinical team evaluating the system. The development therefore depended on three distinct roles: Donald supplied the clinical questions and directed the medical investigation, Brown solved engineering problems and constructed the scanning equipment, and MacVicar helped turn the experimental system into a clinical examination method.

The 1958 Lancet Paper Marked a Clinical Turning Point

On 7 June 1958, Ian Donald, John MacVicar and Tom Brown published Investigation of Abdominal Masses by Pulsed Ultrasound in The Lancet. The title sounds broader and less specifically obstetric than the paper’s later reputation might suggest.

The study demonstrated the use of pulsed ultrasound for abdominal, obstetric and gynaecological investigation and included fetal as well as pelvic images. More importantly for the history of the instrument, the work showed what could be achieved with a compound contact scanner in real clinical examinations.

The images were primitive by modern standards. They were static rather than live video, and early displays lacked the broad range of grey tones associated with later B-mode systems. The operator moved the transducer while the system accumulated information that formed the image on a cathode-ray display.

Yet the clinical value was much easier to see than in a laboratory water tank. The method could be used on patients repeatedly, without an immersion apparatus and without the ionizing radiation used by X-ray imaging.

For obstetric ultrasound, the 1958 paper is therefore a stronger historical boundary than it is for medical ultrasound as a whole. Medical experiments with ultrasound had already been underway for years. What Donald, MacVicar and Brown demonstrated was a workable route toward routine obstetric and gynaecological scanning.

Why Ultrasound Gel Is Part of the Physics, Not a Cosmetic Detail

Direct-contact scanning introduced a practical issue that remains visible in every conventional external ultrasound examination: the transducer must be acoustically coupled to the skin.

Ultrasound does not pass efficiently from a probe through an ordinary air gap into soft tissue. The large difference in acoustic properties causes much of the energy to be reflected at the surface rather than transmitted deeper into the body.

A coupling medium fills microscopic gaps between the transducer and skin and replaces most of that intervening air. This allows more of the emitted sound to enter the body and permits returning echoes to reach the transducer.

The gel used with modern probes is therefore tied directly to the physical problem that early immersion scanners solved with water. The transition from large water tanks to contact scanning did not eliminate acoustic coupling; it replaced a cumbersome form of coupling with one that was far easier to use clinically.

Early B-Mode Images Had to Become Grey-Scale Images

Producing a cross-section did not automatically produce a detailed soft-tissue image. Early B-mode systems often displayed echoes in a relatively limited, almost binary way: a sufficiently strong echo appeared bright while weaker information could be lost.

Grey-scale display changed what could be extracted from the returning signals. Instead of reducing echoes to a narrow bright-or-dark distinction, systems could represent a wider range of amplitudes as different shades of grey.

This made internal texture easier to interpret. Fluid-filled regions, organ boundaries and soft tissues could produce visibly different echo patterns rather than being reduced to a small number of display levels.

The transition depended on improved electronics and display systems as well as more sensitive transducers. During the 1960s and 1970s, grey-scale imaging helped transform ultrasound from an outline-producing device into a method capable of displaying internal soft-tissue detail.

Real-Time Scanning Replaced the Slowly Constructed Static Image

Compound static scanners could produce useful images, but they required the operator to move the transducer mechanically while the system constructed the picture. An examination was therefore fundamentally different from watching anatomy move continuously on a modern display.

The next problem was speed.

The Siemens Vidoson 635, developed by engineers including Richard Soldner and associated with work in Germany during the 1960s, represents an early form of mechanical real-time two-dimensional ultrasound. The Science Museum Group dates its example to 1965 and describes the system as the first ultrasound machine to provide real-time observation inside the body.

The Vidoson rapidly swept its ultrasound beam and generated repeated two-dimensional grey-scale images. In obstetric use, this meant that fetal movement, heartbeat and breathing motion could be observed while the examination was taking place rather than inferred from a slowly assembled static scan.

Real-time imaging changed the relationship between operator and machine. The person holding or positioning the probe could react immediately to the image, alter the scanning plane and follow moving anatomy. Ultrasound was becoming an interactive examination rather than a process that simply produced a picture after a mechanical sweep.

Later electronic arrays reduced dependence on mechanically moving transducers. Multiple transducer elements could be timed electronically so that the ultrasound beam was steered or focused without moving the entire probe assembly. These arrays eventually became standard in modern scanners.

Satomura Added Doppler Information to Medical Ultrasound

Structural imaging answered the question “What is there?” Another branch of ultrasound development addressed “What is moving, and how?”

The physical principle came from Christian Doppler, who described the frequency change associated with relative motion in the nineteenth century. In medical ultrasound, motion between a transducer and a reflecting target can alter the frequency of the returning signal.

Japanese physicist Shigeo Satomura was among the first researchers to apply this effect diagnostically. In the mid-1950s he used ultrasound at approximately 3 MHz to obtain Doppler signals associated with cardiac motion. Work with collaborators, including Ziro Kaneko, later extended the method toward measurements of blood flow in peripheral vessels.

Doppler ultrasound opened a different source of diagnostic information from ordinary B-mode imaging. A conventional structural image locates boundaries and tissues. Doppler methods analyze changes produced by moving reflectors, including blood cells.

Later systems combined these approaches. Duplex ultrasound placed anatomical B-mode information and Doppler measurements within the same examination. Color Doppler then overlaid flow information on the two-dimensional image.

The familiar red and blue display does not simply mean “artery” and “vein.” In a conventional color map, the colors primarily encode the direction of measured motion relative to the transducer according to the selected display settings. An artery can therefore appear in either color depending on beam orientation and the configured map.

Different Medical Specialties Created Different Ultrasound Lineages

By the late 1950s, there was no longer one experimental ultrasound problem. Different medical specialties were selecting different forms of the technology because they needed different information.

In cardiology, the ability to record movement made M-mode valuable before high-quality real-time two-dimensional imaging was available. Later 2D echocardiography and Doppler techniques added chamber anatomy, valve motion and blood-flow information.

In obstetrics and gynaecology, direct-contact B-mode scanning made it possible to examine the uterus, fetus and pelvic masses without the immersion systems used in earlier experimental imaging. Real-time scanning later made fetal movement observable during the examination.

In abdominal and soft-tissue imaging, work by Ludwig, Wild, Reid, Howry and others established that reflected ultrasound could locate structures and distinguish boundaries within tissue. Grey-scale display then allowed much more of the information contained in echo strength to be preserved visually.

The technologies shared acoustic physics, but their clinical histories did not proceed in exactly the same sequence. That is another reason a single-person invention story fails to describe medical ultrasound accurately.

Who Invented Medical Ultrasound?

The most defensible answer depends on what “invented” means in the question. The historical record supports attribution by technical achievement rather than assigning every stage to one person.

Ian Donald therefore should not be described as the sole inventor of medical ultrasound. His place in the history is narrower and better documented: he was one of the central clinical figures in the development of practical obstetric and gynaecological ultrasound, working with MacVicar and engineer Tom Brown.

Likewise, calling Dussik the sole inventor creates a different problem. His experiments preceded the later clinical scanners, but his transmission technique did not become the main architecture of diagnostic ultrasonography, and the anatomical interpretation of his brain images proved unreliable.

Medical ultrasound is better described as a family of technologies assembled through multiple inventions and adaptations.

How the Technology Changed From Experimental Apparatus to Modern Scanner

What Modern Ultrasound Still Inherits From the Early Machines

A current ultrasound system looks very different from Dussik’s water-coupled apparatus, Howry’s immersion tanks, Edler and Hertz’s borrowed reflectoscope or Donald and Brown’s large compound scanners. Much of the underlying measurement logic, however, would be recognizable to the pioneers.

The probe still converts electrical energy into high-frequency mechanical vibration and returning vibration back into electrical signals. Echo timing still provides depth information. Differences between tissues still determine how much sound is transmitted, reflected or attenuated. B-mode still represents echo strength as brightness. Doppler methods still obtain information from motion-dependent changes in the returning signal.

The largest changes have occurred in how quickly and precisely these operations can be performed. Modern arrays contain multiple elements that can be controlled electronically. Digital electronics process enormous numbers of returning echoes, focus beams at different depths, suppress unwanted signals and assemble frames rapidly enough for continuous imaging.

Miniaturization eventually reduced machines that once occupied much of a room to portable carts and hand-carried units. Three-dimensional acquisition, advanced Doppler methods and other later additions expanded the type of information that could be extracted, yet they grew from the same set of problems investigated during the formative decades: how to launch sound into the body, recover usable echoes, determine where those echoes originated and display the result in a form that medicine could interpret.