Physics Today An early history of high-intensity focused ultrasoundWilliam D. O’Brien Jr and Floyd Dunn Citation: Physics Today 68(10), 40 (2015); doi: 10.1063/PT.3.2947 View online: http://dx.doi.org/10.1063/PT.3.2947 View Table of Contents: http://scitation.aip.org/content/aip/magazine/physicstoday/68/10?ver=pdfcov Published by the AIP Publishing

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

98.212.203.54 On: Fri, 09 Oct 2015 00:13:35

40 October 2015 Physics Today www.physicstoday.org

When the Fry brothers—Bill and Frank—arrived at the University of Illinois in1946, their highest priorities were toestablish a laboratory that would de-velop ultrasonic surgical procedures

for the mammalian brain and to study the detailedneuroanatomy of the mammalian central nervoussystem. Within a couple of years, much of the labo-ratory was up and running—animal quarters in-cluded. The faculty and staff of the electrical engi-neering (EE) department were not accustomed tohaving colleagues conduct live-animal studies intheir building. So when the department’s candy ma-chine started to deliver mouse-scarred candy bars,the culprits were rumored to be escapees from thelaboratory cages. The grumbling stopped only afterthe mice were caught and discovered to be wildones—the kind found in the fields of central Illinois.

Two years before the Frys came to the univer-sity, William Everitt had been appointed as head ofthe EE department. He was a renowned communi-cations engineer and was immediately given a leaveof absence until 1 May 1945 to continue his serviceto the country as chief of operational research of theUS Army Signal Corps. When he returned to headthe department, he took it upon himself to developresearch to a high degree.

Everitt was the right person at the right time.When he arrived at the University of Illinois, on-campus engineering research was not common. Atypical attitude voiced by nonresearch faculty was,“If General Electric can’t do it and Westinghouse can’tdo it, we certainly can’t do it.” That attitude reallybothered Everitt, who did indeed succeed in build-ing up significant levels of research.

On 1 September 1949, Everitt became the deanof engineering at Illinois. During the 19 years he heldthat position, he continued his support and encour-agement of high-level research in the College of En-gineering. He was also a founding member of theNational Academy of Engineering.

Cramped quartersJust after World War II ended, colleges and univer-sities across the US experienced a rapid growth intheir faculties, facilities, and enrollments. In partthat was a reaction to the stagnation that had set in during the Depression and war years. But it wasalso fueled by the young men who had returnedfrom wartime service and enrolled in large numbersto take advantage of the GI Bill. Engineering at-tracted many students because the advances anddiscoveries made during the war raised awarenessof the profession. The University of Illinois was noexception to the general trend of rapid growth. In1946 it welcomed one of us (Dunn) as an under-graduate EE student after his return from the Euro-pean theater.

In the immediate aftermath of the war, many inthe scientific and engineering war effort wanted toconduct research of their own choosing in the freeruniversity atmosphere. William J. Fry, shown with

An early history ofhigh-intensity

focused ultrasoundWilliam D. O’Brien Jr and Floyd Dunn

Foresighted leadership, the GI Bill, and research conductedin an unventilated steam tunnel all contributed to the

development of a now ubiquitous medical tool.

Bill O’Brien is the director of the Bioacoustics Research Laboratory and the Donald Biggar Willett Professor Emeritus ofEngineering at the University of Illinois at Urbana-Champaign.Floyd Dunn (1924–2015) joined the electrical engineering faculty at the University of Illinois at Urbana-Champaign in themid 1950s and became an emeritus professor 40 years later.His obituary is in the spring 2015 issue of Acoustics Today.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

98.212.203.54 On: Fri, 09 Oct 2015 00:13:35

his brother in figure 1, was hired by Everitt in late1946, and he established the university’s Bio -acoustics Research Laboratory that same year. Bill,as he was known with affection by colleagues andfriends, had pursued graduate work in theoreticalphysics at the Pennsylvania State University, thesame institution that had granted him a bachelor’sdegree in 1940. After the US entered World War II,he had left his graduate studies to work at the NavalResearch Laboratory in Washington, DC. There he’dspent the war years working to develop principlesfor designing and building sonar systems.

Sonar dates back to French physicist PaulLangevin, who during World War I developed oneof the first uses of ultrasound for underwater echoranging: the detection of submarines, which had beenintroduced as an implement of war. Sonar instru-ments designed between the two world wars werelacking in many respects, though, and before WorldWar II, submarines had never actually been spottedduring hostilities. Thus when Bill left Penn State,there was a crucial need to create better devices.

Francis J. Fry, two years younger than Bill,joined his brother at Illinois in 1946. Like Bill, Frankwent to Penn State, received his bachelor’s degree in 1940, and was hired by Everitt. But unlike hisbrother, Frank had studied EE. After graduation heworked at Westinghouse Electric Corp on power circuit breakers. His division became a prime con-tractor for the Manhattan Project, so Frank spentconsiderable time during World War II at the Radi-ation Laboratory in Berkeley, California, and at theClinton Engineer Works in Oak Ridge, Tennessee.During that time, he also earned his EE master’s de-gree from the University of Pittsburgh.

Space was at a premium in the University ofIllinois EE building when Bill and Frank arrived.Virtually the entire department was crowded into atoo-small, antiquated space that had been expandedand rearranged in every direction as much as pos-sible. The Frys were each assigned a small office, butthey were informed that laboratory space simplydid not exist. However, they did not intend to waituntil space was freed up; a functioning laboratorywas more important than office space. They foundan unoccupied location in a steam tunnel under theEE building that they decided could work. When we

www.physicstoday.org October 2015 Physics Today 41

Figure 1. The Fry brothers, William (left) and Francis (right), joinedthe electrical engineering faculty of the University of Illinois in 1946.

Pioneering ultrasonic

neurosurgery wasconducted by RussellMeyers (second fromleft). Here, Meyers is accompanied byFrancis Fry (left) andWilliam Fry (right), whowere instrumental in developing the techniques thatenabled the operation.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

98.212.203.54 On: Fri, 09 Oct 2015 00:13:35

say “unoccupied,” we mean unoccupied by faculty—the tunnel actually housed many cherished WorldWar I electrical instruments that had been used by the old-time, nonresearch EE faculty. Removingthat equipment for trash disposal did not exactly en-dear the brothers Fry to their faculty colleagues. Asfigure 2 shows, the tunnel was so low that it was im-possible for a person to stand upright. Moreover,during the hot summer months, the unventilatedtunnel became like a steam bath. None of those ob-stacles or inconveniences deterred Bill and Frankfrom setting up their offices and laboratory space.

Everything built from scratch Bill wanted to study the central nervous system socomprehensively that he could begin to understandintimate details of structure and function. At thetime of his arrival at Illinois, the experimental meth-ods in use were rather crude; they required thatphysically rigid electrodes be implanted in brain tissue. The idea was to identify specific anatomicstructures related to brain activity that, for example,caused pain in a particular site of the body. How-ever, the invasive electrode-based procedure pro-duced unreasonably large lesions—damaging ormortal tissue changes.

Bill knew that ultrasound could be focused tovery small volumes. He envisioned it as a vastly su-perior tool to rigid electrodes, one that would allowfor noninvasive alteration of brain tissues. But Billand Frank’s research goals of developing clinicalsurgical procedures and determining a “circuit dia-gram” of the mammalian central nervous systemwould require high-intensity ultrasound: When fo-cused to a pinpoint, HIFU (high-intensity focusedultrasound) could affect deep tissue without harm-ing intervening tissue.

To achieve their unique research goals, the Frybrothers and their coworkers had to invent instru-ments for generating, detecting, and measuring ul-

trasound. Descendants of their inventions are nowused worldwide (see the article by Gail ter Haar,PHYSICS TODAY, December 2001, page 29). The re-searchers also had to understand in detail how ul-trasound propagates in biological media. Their in-vestigations thus addressed such issues as the speedof propagation, how ultrasound is scattered and at-tenuated, and the degree to which tissue can tolerateultrasound. They developed phenomenological the-ories based on their data and even began to studynonlinear acoustic effects, whose understanding isimportant for imaging.

The equipment developed to create, detect, andanalyze ultrasound was built from scratch. Many ofthe materials used today were not available in thosedays. Lack of 21st-century materials was not a prob-lem for the Frys, but lack of 21st-century fundingagencies was a challenge—many of today’s fundingagencies were nascent when the brothers were ini-tiating their research. For example, the Office ofNaval Research was created in 1946 and NSF cameinto existence only in 1950. The National Institute ofHealth (note the singular Institute) had been aroundsince the 1930s, but it largely supported in-house re-search. In the late 1940s, the National Institutes ofHealth (now plural) slowly initiated a program ofgrants for external research. Funding of universityresearch on ultrasound at the time was mostly fromthe military—in particular, the US Navy’s Bureau ofShips and the Office of Naval Research—and grantstended to be for transducer developments. Fortu-nately, Bill was very convincing when he argued onbehalf of the Bioacoustics Research Laboratory’s work.

The Fry brothers began their research journeyby studying how ultrasound affects biological ma-terials. Their specially built, high-intensity, calibratedultrasonic exposure equipment yielded promisingresults from the beginning.1–5 Bill, Frank, and theircolleagues learned, for example, that ultrasound af-fects all kinds of tissue but has different effects ondifferent tissue components. They established thatbelow a certain threshold ultrasound did not injuretissue and that repeated exposures below thatthreshold still did not harm tissue. On the otherhand, at above-threshold intensities, ultrasoundcould affect tissues either temporarily or perma-nently, depending on its amplitude and exposuretime. Much of what they discovered is accepted ascommonplace today, but it was unknown when theFry brothers initiated their studies.

By the mid 1950s, surgical methods based onHIFU were ready for human testing. Before that, theFrys and their coworkers had evaluated their ultra-sonic neurosurgery techniques with cats and mon-keys—more than 250 animals in all. The researchersshowed that their procedures were well tolerated by

42 October 2015 Physics Today www.physicstoday.org

Early ultrasound

Figure 2. A steam tunnel below the University ofIllinois electrical engineering building served as thefirst office and lab space for Bill Fry (seated) and hisbrother, Frank (leaning over). The ceiling was solow that the brothers couldn’t stand up straight.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

98.212.203.54 On: Fri, 09 Oct 2015 00:13:35

the animals and that structures not intended for al-teration by ultrasound were not adversely affected.

Multidisciplinary workTo arrive at the point where the Frys could seri-ously contemplate ultrasound neurosurgery on humans, they had to pull off numerous achieve-ments that spanned the fields of physics, biology,and engineering.‣ The Fry lab designed and built a remarkable newtransducer with a highly confined focal region.6,7 Asfigure 3 shows, the Fry device combined four sepa-rate transducer elements. Each one was fabricatedwith a quartz piezoelectric material; a polystyrenelens positioned in front of the quartz produced a fo-cused field. Individual transducer elements couldproduce high ultrasonic intensities at the focus, buta single transducer would generate a characteristiccigar-shaped focal region. When the four elementsin the Fry design were properly aligned, however,the field in the focal region was localized to a spot.The most intense region of the field could producealterations in biological tissue volumes as small asa few cubic millimeters.‣ Bioacoustics Research Laboratory scientists de-veloped a positioning system that would allow pre-cise movement of the new ultrasonic transducer.8

Great precision was necessary because of the wayultrasound-induced lesions were actually created.One had to move the transducer focus to many spe-cific spots in the brain and produce a very small le-sion at each location. It was the combination of thosemany small, precisely located, spatially adjacent le-sions that yielded the desired complex-shaped lesion.Figure 4 shows the room in which the irradiationswere performed. The 3500-pound positioning sys-tem, mounted on a steel girder frame, was located inan area directly above the irradiation room and con-trolled from the lower room. The Frys electricallyisolated the irradiation room, an effort that demon-strates the care they took to accurately locate the ul-trasonic focus under precisely controlled conditions.‣ Bill, Frank, and colleagues developed and vali-dated the capacity to calibrate ultrasonic fields sothat they could reliably determine field intensities.In those days, there was no National Institute ofStandards and Technology to quantify ultrasound.For that matter, NIST still does not provide such acapability. The Illinois laboratory’s primary calibra-tion standard was based on the radiation force prin-ciple. When ultrasound hits an object, it acts on thebody with a force that can be mathematically relatedto a property of the ultrasound wave, such as poweror intensity. The lab’s primary standard was quiteaccurate and precise, but it was not mobile. So labresearchers fabricated a cleverly designed, portable

thermocouple probe that could be calibrated againstthe primary standard and then serve as a secondarystandard. The thermocouple probe mostly fulfilledthat role in the lab’s animal-based experiments. Ifnecessary, it could also function as a primary stan-dard, provided that the acoustic and thermal prop-erties of the liquid surrounding the thermocouplewere known with accuracy and precision.‣ Illinois researchers determined ultrasonicdosage curves for neural tissue. Such curves relatephysically determined quantities—ultrasonic inten-sity, exposure time, and so forth—to measurable bi-ological outcomes. Figure 5 presents two examples.In one, the biological outcome is paralysis in an an-imal’s hind limb; in the other, the production of abrain lesion.

From movie theater to operating theaterAs noted earlier, the Bioacoustics Research Labora-tory scientists conducted a considerable number ofanimal tests to fully understand HIFU neurosurgeryso that it could be performed on humans. Theyadapted existing techniques and developed new,more precise ones for locating specific anatomicbrain structures. To keep an anesthetized animalperfectly still, they designed a special head holder.A pointer tip connected to the four-beam ultrasonicirradiator (figure 3) designated the location of thebeam’s focus.

Thanks to such innovations, the Frys and theircolleagues could better calibrate the irradiator’s ul-trasound field and position the device so that itsfocus was placed at the desired location in the brain.The 22-minute-long, 16-mm movie Neuro Sonic Sur-gery, produced in 1955, demonstrated all of the ultrasonic neurosurgery procedures that the Frybrothers developed with animals. Dunn was a grad-uate student of Bill’s during the time when many ofthose developments occurred, and he provided thetechnical supervision for the movie.

www.physicstoday.org October 2015 Physics Today 43

Figure 3. A novel transducer designed and built byBill Fry (shown), Frank Fry, and coworkers combinedfour individual transducer elements to create ahighly localized, intense ultrasound spot.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

98.212.203.54 On: Fri, 09 Oct 2015 00:13:35

Meanwhile, Russell Meyers,a professor of neurosurgery atthe medical school of the Univer-sity of Iowa, had gotten to knowthe Frys’ work through their pa-pers and meeting presentationsand had developed an interest inhuman ultrasonic neurosurgery.It was still necessary to developultrasound equipment that couldsafely produce lesions in humans,and the staff at the universityhospital would need years to pre-pare rooms for housing the equip-ment required for human ultra-sonic surgery. But in 1957 Meyerssuccessfully conducted what weunderstand to be the first ultra-sonic surgery on a human brain;the medical milestone was cov-ered in the 2 December 1957 issueof Time magazine. The photo onpage 43 shows Meyers, alongwith the Fry brothers, and one ofhis history-making patients.

In the period 1958–62, some88 patients underwent ultrasonicneurosurgery to treat various af-flictions, including Parkinson’sdisease, cerebral palsy, the aftereffects of stroke, andphantom images and pain following amputation.During that time, the Bioacoustics Research Labora-tory regularly sent personnel to Iowa to oversee andconduct the complex procedures involved in thesurgeries.

Wiring and hiringHuman ultrasonic neurosurgery was the result of atruly synergistic effort that began with the Fry broth-ers’ arrival at Illinois in 1946. The Frys and their col-leagues faced scientific and technical challenges, butthey overcame them, and quite successfully. By theend of the 1950s, HIFU therapy had been born andclinically validated. The Frys had accomplishedtheir original goal of developing ultrasonic surgeryfor affecting the mammalian brain.

After the development of HIFU neurosurgerytechniques, the Bioacoustics Research Laboratory’smain focus turned to mapping brain structures—determining the “wiring” diagram of the central ner -vous system. That was the second of the Frys’ goalswhen they came to Illinois, and they achieved sig-nificant success. However, in large part because ofthe enormous complexity of the media they studied,they could only detail small sections of the cat brain.

The early accomplishments of the BioacousticsResearch Laboratory were only the start. Following

the early neurosonic surgery successes, the uni -versity hired new engineering faculty with biology-related interests; new personnel included Dunn,who joined the EE department as a faculty memberin 1956. Research expanded significantly, but theprincipal focus remained the basic science and ap-plication of the interactions of ultrasound with bio-logical material. As additional faculty came to thelaboratory, it became the de facto center of bioengi-neering in the College of Engineering.

The lab’s real-world-inspired basic science ledto new, medically important ultrasound applica-tions. Lab personnel were the first to quantify in vivohow the ultrasonic absorption coefficient in biolog-ical tissues depends on temperature, and they founda behavior considerably different from that exhib-ited in vitro. Those observations had an importantinfluence on therapeutic ultrasound, and they con-tinue to be valuable, particularly when ultrasonichyperthermia (controlled tissue-temperature increasefrom ultrasound exposure) is considered or used fortreatment of certain types of cancer. In the late 1950sand early 1960s, lab scientists conducted computa-tional research related to ultrasonic imaging. Theirgoal, successfully met, was to develop a computer-controlled HIFU system that included an ultra-sound image to guide therapy.

Bioacoustics Research Laboratory scientists fol-

44 October 2015 Physics Today www.physicstoday.org

Early ultrasound

Figure 4. The heavy, elaborate structure in theupper photo allowed for precise positioning ofthe ultrasonic focus in a room below (lower photo).The positioning system is controlled from below, in the irradiation room.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

98.212.203.54 On: Fri, 09 Oct 2015 00:13:35

lowing up on the brain-lesion studies initiated bythe Frys obtained a fundamental understanding ofthe exposure levels that would just produce irre-versible damage—in other words, they learned toidentify threshold regions separating irreversibletissue damage from nondamage. In the 1970s thoseregions became a de facto worldwide reference fromwhich government and standard-setting bodies es-tablished safety criteria.

These days the fetus of virtually every pregnantwoman in the Western World is evaluated ultrason-ically. Such a widespread use of diagnostic ultra-sound could not have occurred without the safetyassurances that followed from work initially carriedout by a pair of brothers in cramped, hot quartersbelow the plains of central Illinois.

References1. W. J. Fry, V. J. Wulff, D. Tucker, F. J. Fry, J. Acoust. Soc.

Am. 22, 867 (1950).2. W. J. Fry, D. Tucker, F. J. Fry, V. J. Wulff, J. Acoust. Soc.

Am. 23, 364 (1951).3. P. D. Wall, W. J. Fry, R. Stephens, D. Tucker, J. Y.

Lettvin, Science 114, 686 (1951).4. W. J. Fry, J. Acoust. Soc. Am. 24, 412 (1952).5. W. J. Fry, R. B. Fry, J. Acoust. Soc. Am. 25, 6 (1953).6. W. J. Fry, W. H. Mosberg Jr, J. W. Barnard, F. J. Fry,

J. Neurosurg. 11, 471 (1954).7. W. J. Fry, J. W. Barnard, F. J. Fry, R. F. Krumins, J. F.

Brennan, Science 122, 517 (1955).8. W. J. Fry, Neurology 6, 693 (1956).9. F. Dunn, Am. J. Phys. Med. 37, 148 (1958).

10. F. J. Fry, G. Kossoff, R. C. Eggleton, F. Dunn, J. Acoust.Soc. Am. 48, 1413 (1970). ■

Figure 5. Dosage curves relate physical properties of applied ultrasound to a biological effect. (a) The curves here indicate, foranesthetized mice prepared at various temperatures T, the time (t) and intensity at which 50% of the subjects exposed to 982-kHzultrasound experienced paralysis in a hind limb. Also indicated, near the data points expressed as circles, are maximum observedtemperature rises. (Adapted from ref. 9.) (b) Three distinct studies, including one at the University of Illinois, determined the intensity and exposure times needed to create brain lesions in mammalian subjects. The numbers on the data points indicate ultrasound frequency in megahertz. (Adapted from ref. 10.)

Coming January 2016A whole new look for Physics Today

ACOUSTIC INTENSITY / (w/cm ) /1 2 21 2

0

0.2

0.4

0.6

0.8

1.0

1.2

6 7 8 9 10 11 12 13

11.0 °C

4.5 °C

6.2 °C

6.2 °C

4.8 °C

6.1 °C

7.7 °C

9.3 °C

9.0 °C

7.1 °C

16.5 °C

5.7 °C5.7 °C

5.0 °C

T=

2°C

T=

20

°C

T=

10

°C

t−1

−1

(s)

AC

OU

ST

IC I

NT

EN

SIT

Y(w

/cm

)2

10110−110−210−3

103

102

104

105

10−4

TIME (s)

Illinois

Pond

Lele

2.7

31

1

1

1

11

3

3

3

1

1

2

3

3

3

3

3

5

64

1

a

b

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

98.212.203.54 On: Fri, 09 Oct 2015 00:13:35