ORIGINAL ARTICLE: Clinical Endoscopy

First clinical trial of the ‘‘MiRo’’ capsule endoscope by using

a novel transmission technology: electric-field propagation
Seungmin Bang, MD, Jeong Youp Park, MD, Seok Jeong, MD, Young Ho Kim, PhD, Han Bo Shim, PhD,
Tae Song Kim, PhD, Don Haeng Lee, MD, PhD, Si Young Song, MD, PhD
Seoul, Korea

Background: We developed a capsule endoscope (CE), ‘‘MiRo,’’ with the novel transmission technology of
electric-field propagation. The technology uses the human body as a conductive medium for data transmission.
Specifications of the prototype include the ability to receive real-time images; size, 10.8  24 mm; weight, 3.3 g;
field of view, 150 ; resolution of power, 320  320 pixels; and transmittal speed, 2 frames per second.
Objective: To evaluate the clinical safety and diagnostic feasibility of the prototype MiRo, we conducted
a multicenter clinical trial.
Design and Patients: All volunteers underwent baseline examinations, including EGD and electrocardiography
for the screening of GI obstructive and cardiovascular diseases, before the trial. In the first 10 cases, 24-hour
Holter monitoring was also performed. To evaluate the diagnostic feasibility, transmission rate of the captured
images, inspection rate of the entire small bowel, and quality of transmitted images (graded as outstanding,
excellent, good/average, below average, and poor) were analyzed.
Results: Of the 49 healthy volunteers, 45 were included in the trial, and 4 were excluded because of baseline
abnormalities. No adverse effects were noted. All CEs were expelled within 2 days, and the entire small bowel
could be explored in all cases. The transmission rates of the captured image in the stomach, small bowel, and
colon were 99.5%, 99.6%, and 97.2%, respectively. The mean total duration of image transmission was 9 hours,
51 minutes, and the mean transit time of the entire small bowel was 4 hours, 33 minutes. Image quality was
graded as good or better in 41 cases (91.1%). Details of the villi and vascular structures of the entire small bowel
were clearly visualized in 31 cases (68.9%).
Conclusions: MiRo is safe and effective for exploring the entire small bowel, with good image quality and real-
time feasibility. This novel transmission technology may have applications beyond the field of capsule endos-
copy. (Gastrointest Endosc 2009;69:253-9.)

Since the first development of an innovative wireless Furthermore, comfortable evaluation of the entire colon
capsule endoscope (CE) in 2000, the fundamental para- is now possible with more recently developed CEs.4,5
digm for the diagnosis of small-bowel diseases has The first commercialized CE, M2A (Given Imaging,
changed.1-3 The wireless CE allows for the exploration of Yoqneam, Israel), was developed by scientists in Israel
the entire small bowel without pain or discomfort. who were able to miniaturize the electronic components
needed to transmit video signals to an external receiver
Abbreviations: CE, capsule endoscope; CMOS, complementary metal by using telemetry technology. The basic meaning of te-
oxide silicon; ECG, electrocardiography; LED, light-emitting diode; lemetry is carrying information from one place to another
RF, radiofrequency.
through some medium or channel. The typical communi-
DISCLOSURE: All authors disclosed no financial relationships relevant to cation system is composed of 3 elementary components.
this publication. This multicenter trial was supported by the 21st Century The transmitter, at some location in space, converts the
Frontier R&D Project (F0001004-2006-31), sponsored by the Korean
message signal produced by a source of information into
Ministry of Commerce, Industry, and Energy, and by Intromedic.
a form suitable for transmission over a channel. The chan-
See CME section; p. 303.
Copyright ª 2009 by the American Society for Gastrointestinal Endoscopy
nel, in turn, transports the message signal and delivers it
0016-5107/$36.00 to a receiver at some other location in space.6 The state-
doi:10.1016/j.gie.2008.04.033 of-the-art CE system can be interpreted as an application

www.giejournal.org Volume 69, No. 2 : 2009 GASTROINTESTINAL ENDOSCOPY 253

Clinical trial of the novel capsule endoscope ‘‘MiRo’’ Bang et al

of a generic communication system. The CE system has no

direct signal path, such as a conductive wire, between the Capsule Summary
capsule (information source) and data recorder. The sim-
plest means of data transmission from this noninvasive in What is already known on this topic
vivo device to the receiver is the use of radiofrequency d Available capsule endoscopy equipment uses
(RF) communication technology. Until now, CEs available radiofrequency communication technology for image
in the clinical field only used RF communication technol- transmission.
ogy for data transmission.
We developed another CE system based on a novel hu- What this study adds to our knowledge
man telemetry technology known as electric-field propa- d Capsule endoscopy by using electric-field propagation
gation, which is now patented in the United States.7 Our transmission technology in 45 healthy individuals
novel human body communication technology uses the resulted in captured image rates in the stomach, small
human body as a conductive medium for data transmis- bowel, and colon of 99.5%, 99.6%, and 97.2%,
sion. The new technology made it possible to finalize respectively.
d Image quality was graded as good or better in 91.1%,
the physical structure of the capsule with very simple and the villi and vascular structures of the entire small
transmitters, as a pair of gold plates coated on the surface bowel were clearly visualized in 68.9%.
of the capsule. This dramatically reduces the power con-
sumed compared with existing communication devices
that use RF transmission technology, which contains neuromuscular morbidities, eg, muscular dystrophy, were
power-consuming components, such as a local oscillator also excluded.
that makes a very-high-frequency carrier, an amplifier Volunteers ingested MiRo after taking 4 L of polyethyl-
that lets signal transport high power as well as loss of ene glycol solution and fasting for 8 hours. Just before
most energy generated from transmitter because of radia- this, an array of 8 receivers was attached to the abdominal
tion characteristics of RF energy. In addition to a longer wall of each subject, and a recorder attached to a belt was
operation time, our CE takes advantage of the surplus worn around the waist. Volunteers were encouraged to
energy to produce more image data. continue their usual activities, and, after 24 hours, the
We conducted the first multicenter clinical trial with the recorders were disconnected and the receivers were re-
prototype by using healthy volunteers. This study was moved. The recorded information was uploaded to a com-
aimed at verifying the safety of the MiRo capsule (Intro- puter, and the images captured by MiRo were analyzed by
Medic Co, Ltd, Seoul, Korea) in human beings, especially using proprietary MiroAppl software. Subjects were asked
with regard to the cardiac and neuromuscular systems, to look for MiRo in their feces. All 45 capsules were re-
and evaluating the validity of MiRo in the diagnosis of trieved between 12 hours to 2 days after ingestion. Analy-
human small-bowel diseases. sis of the recorded images was performed by 3 capsule
endoscopy experts who had experience with more than
50 cases that involved CEs.
PATIENTS AND METHODS
Specifications of CE prototype
Study design The prototype capsule, MiRo, was 10.8  24 mm,
Healthy volunteers (26 men, 23 women; aged 23–57 weighed 3.3 g, had a field of view of 150 , resolution
years; mean age 38.9 years) were enrolled at 2 hospitals power of 320  320 pixels, and a battery life of 9 to 11
in Korea. All volunteers gave informed consent according hours in preclinical tests (Fig. 1). A white-light–emitting di-
to the requirements of the local and government ethics ode (LED) was used as the illumination source, and the
committees. This clinical trial was approved by the Korean optical system focused reflected optical rays onto a com-
Food and Drug Administration and by the institutional plementary metal oxide silicon (CMOS) sensor. The sensi-
review boards at the participating hospitals. tive, low-power CMOS image sensor converted the optical
Before enrollment, we performed screening tests for rays to electrical voltages. The capsule uses 2 serial silver
health problems, including routine electrocardiography oxide batteries as a power source. Only one chip is used
(ECG), chest and abdomen radiographs, EGD, and blood in the capsule, which includes the CMOS image sensor
chemistry tests. Exclusion criteria included age (!19 years and a parallel-to-serial data converting logic block. The se-
old and O70 years old), pregnant or breastfeeding status, rial digital signal is limited in power to a safe level in the
GI diseases (obstructive diseases, malignancies, multiple human body and propagates as a low-power electric field
diverticula), and severe morbidities, eg, liver cirrhosis. Pa- outside of the body via 2 gold plating bands on the hous-
tients with cardiac arrhythmias with or without cardiac ing. The capsule transmitted continuous video images at 2
pacemakers or implantable cardiac defibrillators and those frames/s during passage through the GI tract. Images were
with ischemic heart disease and hypertension, history of recorded by using a solid-state recorder and an aerial
stroke, seizure disorders, previous brain surgery, or other system applied to the skin of the body.

254 GASTROINTESTINAL ENDOSCOPY Volume 69, No. 2 : 2009 www.giejournal.org

Bang et al Clinical trial of the novel capsule endoscope ‘‘MiRo’’

and choose one of the results. The opinion of the third

expert was asked in 11 of 45 cases. The quality of the
captured images was graded as outstanding, excellent,
good/average, below average, or poor. To assess image
quality, the brightness, contrast, and degree of resolution
of the villi and vasculature were considered. ‘‘Outstand-
ing’’ indicated that the brightness and contrast were har-
monious and the delineation of the villi and vasculature
was clear in more than 90% of the entire small bowel.
‘‘Excellent’’ meant that the parameters were fulfilled in
75% to 90% of the entire small bowel, ‘‘good/average’’
fell between 60% to 75%, ‘‘below average’’ between 50%
to 60%, and ‘‘poor’’ in less than 50% of the entire small
bowel.
Technical factors, including the success rate of the en-
tire small-bowel exploration, rate of successful image
transmission, total duration of image transmission, and ef-
fective battery operation time, were also analyzed. We also
sought to determine whether the MiRo system suffered in-
terference from electric and/or electronic devices encoun-
tered in daily life, such as televisions, mobile phones, and
Figure 1. A, Schematic diagram of MiRo: optical dome (1) lens (2), LEDs
the subway.
(3), image sensor (4), battery (5). B, Photograph of MiRo.

RESULTS
Analysis of safety
To verify the safety of MiRo, we evaluated any adverse After screening the subjects, 4 subjects were dropped
effects on the GI tract and the cardiovascular and neuro- from the trial. One was diagnosed as having liver cirrhosis
muscular systems. To assess effects on the GI system, with esophageal varices, two showed ECG abnormalities,
any abnormal symptoms, including abdominal pain or and one withdrew consent before swallowing the capsule.
cramps, were recorded by volunteers. To assess effects Thus, a total of 45 volunteers participated. The mean age
of MiRo on the cardiac conduction system, 10 subjects un- was 38.9 years (23–57 years), and 26 were men.
derwent 24-hour Holter monitoring during the GI passage
of MiRo. All volunteers were checked by 12-lead and Safety of MiRo
rhythm strip ECG before ingestion of MiRo and during One subject recorded chest pain during the test, and
the test. Any major abnormality in cardiac rhythm, such the testing capsule was expelled. During the test, one sub-
as tachycardia, bradycardia, fibrillation, or symptomatic ject complained of a tingling sensation in the right fore-
ventricular premature beat, was considered to be an ad- arm and another described back pain. However, the 2
verse effect from MiRo. Elevation of blood pressure by volunteers had experienced the same episodes intermit-
more than 20 mm Hg from baseline and symptomatic hy- tently before the test. One recorded right upper-quadrant
potension were considered adverse events, as was change pain during the test, but, by that time, the capsule was in
in heart rate of more than 30% from baseline. Difficulty in the distal rectum (Table 1).
performing daily activities, such as writing, typing, or driv-
ing, and the development of any involuntary movement, Diagnostic feasibility
including tremor and convulsions, were monitored during The mean operation time of MiRo was 9 hours, 51 min-
the test to assess adverse effects on the neuromuscular utes (5 hours, 35 minutes to 11 hours). Complete explora-
system. tion of the entire small bowel was achieved in all 45
volunteers. Furthermore, exploration of the large bowel,
Analysis of diagnostic feasibility at least the proximal segment, was possible in all cases.
Diagnostic feasibility was analyzed in terms of the qual- In terms of image transmission rate, MiRo successfully
ity of images captured by MiRo and transit time in each or- transmitted 99.5%, 99.6%, and 97.2% of captured images
gan from the esophagus to the colon. Three CE experts in the stomach, small bowel, and large bowel, respectively.
who were blinded to the others’ findings analyzed the The transmission rate in the esophagus was 87.4%
videos. Two CE experts first analyzed the videos and re- (Table 2). Most volunteers ingested the capsule in the sit-
ported their results. If the two results were not in agree- ting or standing position. In these positions, the capsule
ment, the third expert was asked to review the videos passed through the esophagus so rapidly that we were

www.giejournal.org Volume 69, No. 2 : 2009 GASTROINTESTINAL ENDOSCOPY 255

Clinical trial of the novel capsule endoscope ‘‘MiRo’’ Bang et al

TABLE 1. Safety of MiRo in the human body TABLE 2. Technical outcome of diagnostic feasibility
of MiRo
Change in cardiac conduction system
Rate of entire small-bowel 45/45
Change in 24-h Holter monitoring 0/10 exploration
Change in 12-lead ECG 0/45 Mean total duration of image 9 h, 51 min
Change in heart rate* 0/45 transmission, (range) (5 h, 35 min to 11 h)

Change in blood pressurey 0/45 Rate of image transmission,

% (range)
Development of cardiac-origin chest pain 0/45
Esophagus 87.4 (0–100)
Development of abdominal painz 1/45
Stomach 99.5 (96.4–99.9)
Development of involuntary movement 0/45
Small bowel 99.6 (97–99.6)
Disturbance in performing daily tasks 0/45
Large bowel 97.2 (0–99.9)
Otherx 2/45
*Change of more than 30% from baseline heart rate was considered
abnormal.
yAn increment of R20 mm Hg from baseline or R150/100 mm Hg
or symptomatic hypotension was considered abnormal.
endoscopic images.9-11 In addition, currently available
zOne subject complained of right upper-quadrant pain when the CEs only take images while being passively moved
capsule was almost in the distal rectum. through the intestinal tract by peristalsis.
xOne subject complained of right forearm tingling and another of
back pain; both were irrelevant to the test.
To overcome such limitations, CEs are being modified
to improve battery life, illumination, optics, angle of
view, etc. To make a better CE, we developed a more intel-
unable to obtain good images in some cases. The capsules ligent capsule endoscopy system based on electric-field
were safely expelled from the anus in all cases within 2 propagation in cooperation with the Intelligent Microsys-
days. Daily activity that involved electric and/or electronic tem Center of Korean Institute of Science and Technology,
devices, such as watching television and taking the IntroMedic Co, Ltd, and Yonsei University College of Med-
subway, did not interfere with CE performance. icine under the Korean Frontier 21 project. The main dif-
In 31 cases (68.9%), the images were fine and sophisti- ference between the MiRo and conventional CEs is that
cated, and revealed microstructures over more than 75% MiRo uses the human body itself as the communication
of the entire small bowel (Fig. 2). In only 4 cases were channel. The standard capsule endoscopy system uses
the images below average quality, which was attributable an RF module to send information from the image sensor
to poor bowel preparation, with a great deal of free-float- through living tissue to the outside of the human body, re-
ing food and bubbles present. The mean transit time in garding the human body as a nonconductor. However, the
the stomach was 41 minutes, 40 seconds and 4 hours, ionic materials and free electrons block the signal and at-
33 minutes in the small bowel (Table 3). tenuate their energy by consuming the electric field with
We found 13 incidental lesions in the small bowel and 1 drift current. In contrast, MiRo uses the human body as
polyp in the colon (Fig. 3). Among the 13 lesions in the a semiconductor in which an electric field can be induced
small bowel, 4 cases of small-bowel polyps, 5 of lymphan- and consequently generates drift current, even though the
giectasia, and 4 of lymphoid hyperplasia were suspected. body has poor conductivity compared with a metal wire.
However, these lesions were not found in the thin barium, As a result, MiRo does not require an additional antenna
double-contrast small-bowel radiographic series. or high-frequency circuit for remote communication. No
need of space inside the capsule for an antenna or RF
module enables the MiRo to have more room available on-
DISCUSSION board, perhaps enabling other functions, such as biopsies
or locomotive guidance in the future, or allowing the size
Since the introduction of capsule endoscopy in GI-tract to be reduced.
diagnostics, the paradigm for choosing diagnostic tools in With this conceptual difference, MiRo has 2 notable ad-
GI disorders has changed.1-3 Exploration of the small vantages when compared with conventional CEs. First, it
bowel is now possible without discomfort to patients or consumes less power because it does not perform the
without the need for a well-trained endoscopist. However, high-frequency modulation process and antenna driving,
conventional CE has had important limitations, such as which allows the battery to last longer. The use of power
the duration of image capture in the human body.8 Fur- was reduced in 2 additional ways. The CMOS image sensor
thermore, images captured in the GI tract have not shown was designed to minimize power consumption, and the te-
sufficiently fine resolution compared with conventional lemetry chip and image sensor were combined on one chip

256 GASTROINTESTINAL ENDOSCOPY Volume 69, No. 2 : 2009 www.giejournal.org

Bang et al Clinical trial of the novel capsule endoscope ‘‘MiRo’’

Figure 2. Samples of images taken by MiRo that were scored as outstanding. A, Esophagogastric junction. B, Antrum of stomach. C and D, Normal villi
and vasculature of the small bowel. E and F, Normal colon.

to reduce the current required for fan out between chips.12 the CE to explore the entire small bowel; however, MiRo
As noted above, the mean transit time of CE from the could explore the entire small bowel even in cases in which
mouth to the end of small bowel was almost 6 hours, small-bowel transit time was more than 8 hours.
and it took more than 8 hours in some cases in our trial. A major limitation of currently available CEs is image qual-
This variable transit time in the GI tract does not allow ity, which is inferior to that of conventional endoscopes.8-11

www.giejournal.org Volume 69, No. 2 : 2009 GASTROINTESTINAL ENDOSCOPY 257

Clinical trial of the novel capsule endoscope ‘‘MiRo’’ Bang et al

TABLE 3. Clinical outcomes in diagnostic feasibility

of MiRo

Quality of image
(no. [%])
Outstanding 17 (37.8)

Excellent 14 (31.1)
Good/average 10 (22.2)
Below average 4 (8.9)
Poor 0
Mean bowel transit
time of MiRo (range)
Esophagus 6 min, 47 s (0 to 2 h, 24 min, 26 s)
Stomach 41 min, 40 s (1 min, 29 s to 3 h, 36
min, 44 s)
Small bowel 4 h, 33 min, 22 s (1 h, 15 min, 12 s
to 8 h, 11 min, 4 s)
Rate of expulsion of 45/45 (100%)
capsule via the anus

There are several reasons for inferior image quality, eg,

data compression. Conventional CEs perform image
data compression to efficiently reduce the number of
bits and forward-error correction coding but this leads
to lower image quality, especially at high frequencies.
As a result, data compression causes blurring at the
edges of objects and of small or thin objects. However, Figure 3. A, Polypoid lesions found in the jejunum. B, Polypoid lesion in
MiRo does not require data compression, which results the proximal colon.
in more precise images. Improved image quality when
using MiRo was confirmed in this study. As mentioned
above, the fine structures of the bowel mucosal surface, onstrate whether this technologic advantage is translated
including villi and vasculature of the entire small-bowel into clinically better CE. We also expect a randomized
lumen, could be observed without blurring or distortion comparison study among the upgraded MiRo, CEs, and
in more than 90% of cases. The electric-field propagation improved Given PillCam CE SB2 in the future.
theory used in MiRo induces an electrical field and cur-
rent drift in the human body, thus electrophysiologic
safety in the cardiac conduction and neuromuscular sys-
tems was a focus of the current study. We observed no REFERENCES
adverse effects in these systems as a result of MiRo. In 1. Iddan G, Meron G, Glukhovsky A, et al. Wireless capsule endoscopy.
this trial, the unique concept of MiRo, electric-field prop- Nature 2000;405:417.
agation, was demonstrated to be safe in the human body. 2. Swain CP, Goong F, Mills TN. Wireless transmission of a color television
In summary, the MiRo has advantages over conven- moving image from the stomach using a miniature CCD camera, light
source, and microwave transmitter [abstract]. Gut 1996;39:A26.
tional CEs in terms of duration of operation, completeness
3. Meron GD. The development of the swallowable video-capsule (M2A).
of data transmission, and quality of images. The safety of Gastrointest Endosc 2000;52:817-9.
MiRo was verified in this study, and it has the potential 4. Eliakim R, Fireman Z, Gralnek IM, et al. Evaluation of the PillCam Colon
to attach more intelligent systems, including biopsies, be- capsule in the detection of colonic pathology: results of the first mul-
cause of the device’s structure. In addition, the prototype ticenter, prospective, comparative study. Endoscopy 2006;38:963-70.
5. Schoofs N, Devière J, Van Gossum A. PillCam colon capsule endoscopy
has been modified to a specification of 3 frames/s, with
compared with colonoscopy for colorectal tumor diagnosis: a prospec-
a longer operating time. The upgraded MiRo received tive pilot study. Endoscopy 2006;38:971-7.
a certificate from the Korean Food and Drug Administra- 6. Haykin S, Moher M. Introduction to analog and digital communica-
tion and a CE Mark. More clinical trials are needed to dem- tions. 2nd ed. New Jersey: Wiley; 2007.

258 GASTROINTESTINAL ENDOSCOPY Volume 69, No. 2 : 2009 www.giejournal.org

Bang et al Clinical trial of the novel capsule endoscope ‘‘MiRo’’

7. Kim TS, Park JO, Moon SW, et al, inventors; Korea Institute of Science 12. Kim TS, Song SY, Jung H, et al. Micro capsule endoscope for gastro in-
and Technology, assignee, Method and apparatus for communication testinal tract. Conf Proc IEEE Eng Med Biol Soc 2007;2007:2823-6.
between inside and outside of transmission medium using transmis-
sion medium as communication line. US patent US 7,307,544 B2.
December 11, 2007.
8. Swain P, Fritscher-Ravens A. Role of video endoscopy in managing Received February 5, 2008. Accepted April 12, 2008.
small bowel disease. Gut 2004;53:1866-75. Current affiliations: Division of Gastroenterology (S.B., J.Y.P., S.Y.S.),
9. Saurin JC, Delvaux M, Gaudin JL, et al. Diagnostic value of endo- Department of Internal Medicine, Yonsei Institute of Gastroenterology,
scopic capsule in patients with obscure digestive bleeding: blinded Yonsei University of College of Medicine, Seoul, Department of Internal
comparison with video push enteroscopy. Endoscopy 2003;35: Medicine (S.J., D.H.L.), Inha University, Incheon (Y.H.K., H.B.S.), IntroMedic
576-84. Co, Ltd, Seoul, Intelligent Microsystem Center (T.S.K.), Korean Institute of
10. Mylonaki M, Fritscher-Ravens A, Swain P. Wireless capsule endos- Science and Technology, Seoul, Korea.
copy: a comparison with push enteroscopy in patients with gas-
troscopy and colonoscopy negative gastrointestinal bleeding. Gut Reprint requests: Si Young Song, MD, PhD, Division of Gastroenterology,
2003;52:1122-6. Department of Internal Medicine, Institute of Gastroenterology, Yonsei
11. Appleyard M, Fireman Z, Glukhovsky A, et al. A randomized trial University College of Medicine, Shinchon-dong 134, Seodaemoon-gu,
Seoul, Korea, 120-752.
comparing wireless capsule endoscopy with push enteroscopy for
the detection of small-bowel lesions. Gastroenterology 2000;119: If you want to chat with an author of this article, you may contact him at
1431-8. [email protected].

www.giejournal.org Volume 69, No. 2 : 2009 GASTROINTESTINAL ENDOSCOPY 259