- Email: [email protected]
Usability and performance of a wearable tele-echography robot for focused assessment of trauma using sonography
Medical Engineering & Physics 35 (2013) 165–171
Contents lists available at SciVerse ScienceDirect
Medical Engineering & Physics journal homepage: www.elsevier.com/locate/medengphy
Usability and performance of a wearable tele-echography robot for focused assessment of trauma using sonography Keiichiro Ito a,∗ , Shigeki Sugano a , Ryohei Takeuchi b , Kyota Nakamura c , Hiroyasu Iwata d a
Department of Creative Science and Engineering, Waseda University, Japan Department of Joint Surgery Center, Yokosuka Municipal Hospital, Japan Department of Emergency Medicine, Yokohama City University Hospital, Japan d Waseda Institute for Advanced Study (WIAS), Waseda University, Japan b c
a r t i c l e
i n f o
Article history: Received 6 January 2012 Received in revised form 27 April 2012 Accepted 28 April 2012 Keywords: Emergency medicine Tele-echography Wearable system
a b s t r a c t Focused Assessment with Sonography for Trauma (FAST) is widely used as a first lifesaving step for patients suffering from internal bleeding. Because it may take a long time to transport such patients to a hospital, a wearable and portable tele-echography robot that a paramedic can attach to the patient has been developed. In the current study, experiments were conducted to evaluate the usability and performance of attached FAST. The proposed robot must be attached to 4 areas to perform FAST. The time required for attachment and the positions of attachment completed by 9 non-medical staff members, as well as the time it took for the FAST to reach a medical doctor, were measured. The echo images obtained when the patient’s body was in motion were evaluated by a medical doctor. The robot could be attached to all 4 areas within approximately 5 min, and the maximum gap was 4.8 cm. This indicates that a paramedic who has received training in emergency medical care should be able to attach the robot to a patient quickly and accurately. Additionally, it was confirmed that the robot could be used to complete FAST under a doctor’s control within 9 min and that the extracted echo images were suitable for FAST. A comparison of the results with current ambulance transportation time confirmed that FAST could be completed approximately 14 min before the patient reached the hospital. The results of the current study indicate that the robot is worth using, is suitable for FAST, and will be effective in emergency medical care. © 2012 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Echography is a critical diagnostic approach in the treatment of shock patients who are severely injured in traffic accidents or large-scale disasters, and Focused Assessment with Sonography for Trauma (FAST) is a quick, easy, and effective echography method that is widely used in the initial hospital evaluation of traumatic shock patients [1–7]. Because transporting a patient from an injury site to a hospital often takes a long time, a tele-echography system that a paramedic can easily use at the injury site or in the ambulance should increase the patient’s likelihood of surviving. The system’s portability and ease of use by paramedics are very important. The system should be small enough to be easily used in an ambulance,
∗ Corresponding author at: Department of Creative Science and Engineering, School of Modern Mechanical Engineering, Waseda University, 17 Kikui-cho, Shinjuku-ku, Tokyo 162-0044, Japan. Tel.: +81 3 3203 4382; fax: +81 3 3203 4382. E-mail address: [email protected] (K. Ito).
and it should be easy to handle, allowing a paramedic to use it quickly and properly. Many tele-echography systems have been developed recently. Zhu et al., for example, developed a robotic system for carotid artery tele-echography [8]. Vilchis et al. developed a tele-echography system in which a physician adjusts the location of an echo probe moved by a cable-controlled robot strapped to the body of a patient at a distant site [9,10]. Masuda et al. developed a wirelessly controlled system with an ultrasound (US) probe mounted on the arm of a pantograph mechanism attached to the side of the patient’s bed [11,12]. Finally, through the field testing of a mobile robotic teleecho system placed in an ambulance, Takeuchi et al. have confirmed that this system could be remotely operated and send clear, realtime US images of a patient’s abdomen to the destination hospital [13]. These systems, however, are hard to carry and require a lot of bedside space. Robosoft Inc. therefore developed a portable teleechography robotic system [14] that is pushed against the patient’s body by a medical assistant while a physician remotely controls the position of the US probe. Although the system is easy to transport,
1350-4533/$ – see front matter © 2012 IPEM. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.medengphy.2012.04.011
166
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
it is not easy for medical assistants to use because they have to determine the location on the patient’s body to apply pressure to activate the system by interpreting the often vague instructions of a doctor, and they must apply constant pressure. The purpose of the work presented here was to experimentally evaluate the usability and performance of a prototype wearable tele-echography robot designed to be used easily by a paramedic in an ambulance. 2. Materials and methods
site. Information concerning the position and orientation of the US probe, as well as image and voice information, is also transmitted. The doctor performs the tele-echography while observing the echo image, the US probe, and the patient. The portable echo device consists of a MicroMaxx (SonoSite Inc.) 30 cm wide, 27 cm long, and 8 cm high (weight: 3.9 kg) and a sector US probe (SonoSite Inc., P17/5-1). The frequency range of the US probe used for the chest and abdomen is 1–5 MHz. A portable battery for the system is also required. In addition, different types of networks are required for the system: ISDN (typically 128 kb/s), LAN, and WiMAX (IEEE 802.16e, download: 40 Mb/s, upload: 10 Mb/s).
2.1. Overview of the wearable tele-echography robot system 2.2. Wearable tele-echography robot An overview of the wearable tele-echography robot system is shown in Fig. 1(a). A medical doctor operates the graphical user interface on the computer, and the control signals are transmitted through a network to the robot in an ambulance or an injury
The robot can be easily attached to the trunk of the patient’s body using a corset belt, and it naturally adapts to the patient’s body motion. It has a control mechanism that enables it to perform
Fig. 1. Proposed wearable tele-echography robotic system. (a) System overview. To control the wearable tele-echography robot and transmit audio and camera data, a mobile network will be used for the connection between the hospital and the patient. The system fits into a suitcase 27 cm wide, 50 cm long, and 70 cm high. (b) Wearable teleechography robot. The robot provides 4 degrees of freedom (pitching, rolling, positioning, and contacting) and control of the US probe for FAST. Additionally, the robot has two curvature rails, a soft urethane sponge, an elastic silicon-based corset, two rotary motors, a linear motor, and two mechanical springs. It can be attached to a patient by wrapping the corset around the patient’s trunk. The probe is kept in contact with the patient’s skin by two mechanical springs. The robot measures 200 mm × 200 mm × 100 mm and weighs 2.2 kg.
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
echography, and it can generate a force that keeps the US probe in contact with the patient’s body. As shown in Fig. 1(b), the robot has 4 degrees of freedom: pitching, rolling, positioning, and contacting. The curvature rails and rotary motors (Harmonic drive, RSF-5A, 66 g) are used to produce the pitching and rolling motions of the probe: 90 degrees pitching and 360 degrees rolling. A linear motor (Hitachi, S080Q, weight; 80 g) is used to position the probe and is able move it as much as 100 mm. The urethane sponge is used as a cushion between the robot and the patient’s skin, and the two mechanical springs (Samini Inc., Compression spring, Coefficient Kf; 0.27 N/mm) push the US probe against the patient. In designing the probe control mechanism for echography, the main design objectives were to make the mechanism simple and to make the robot small. Because the US probe needs to be in contact with the body surface even during pitching, rolling, and motion across the body surface, a mechanism that generates contact force passively with two mechanical springs was developed. The contact force is generated by attaching the robot; the probe is pushed onto the surface of the body. Additionally, the proposed mechanism keeps the point of the probe in contact with the patient’s skin, and a contact force is continuously generated when the probe pitches, rolls, and moves across the body from springs installed in the end of the robot. Koizumi et al. reported that the contact force of the US probe should be no more than 9 N so that the area of diagnosis is not pressed
167
excessively by the probe and the echo image is not distorted [15]. The stroke of the two springs was therefore set to 10 mm. When the robot is attached to the patient, this mechanism can generate a contact force of approximately 5.4 N. The usability and performance of the system for FAST were evaluated in 2 experiments: an attachment experiment using examinees and a remote-controlled FAST experiment. Because FAST narrows the area of diagnosis down to 4 parts—pericardium, liver, spleen, and bladder—the attachment time and position in all 4 areas were used as indicators of the usability of the system. In this experiment, examinees were divided into 2 groups: examinees attaching the robot and examinees having the robot attached to them. The patient’s body type is a key indicator for how the robot should be attached. The examinees having the robot attached to them were classified into 3 types (slender, normal, and overweight) based on their BMIs (BMI: 18, 22, and 28, respectively). Because the skill level of actual paramedics would potentially affect the result of the experiment, the minimum usability of the system was evaluated using 9 non-medical staff members who had the same low level of skill. The examinees attached the robot to all 4 areas while looking at a figure of attachment positions and procedures (Fig. 2). Attaching the reference lines was based on the line through the umbilicus and the middle of the sternum and the line through the nipples. The attachment times, starting when the robot was picked up and ending when the robot had been attached to each
Fig. 2. Attachment position and attaching procedure. The examinees attached the robot on each body type, without instructions from the doctor, by looking at the attachment position and the pictures showing the procedure. The reference lines for attachment were based on the line between the nipples (for FAST (Focused Assessment with Sonography for Trauma) areas 1, 2, and 3) and the line between the umbilicus and the center of the sternum (for FAST area 4); thus, even non-medical personnel could easily understand the attachment positions. The examinees attached the robot along the reference lines. The best attachment positions of the robot were assumed to be the areas where FAST can be performed by the doctor’s remote control. The measurement of the gaps between the attachment positions and FAST areas 1 and 4 were based on the umbilicus–sternum line, and the measurement of the gaps between the attachment positions and FAST areas 2 and 3 were based on nipple–nipple line.
168
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
Fig. 3. FAST (Focused Assessment with Sonography for Trauma) experiment. The robot was attached to FAST 1 area based on the result of the attachment experiment, which was defined as the maximum gap. The doctor remotely controlled the robot via a wired local network, and the time required to extract the echo image for FAST area 1 (pericardium) was measured. The control of the US (ultrasound) probe and the extraction of the echo image were conducted in the same way on each FAST area, and times required to obtain FAST images for each area were also measured. The doctor evaluated the sharpness of the echo images obtained for each FAST area with regard to their suitability for FAST.
FAST area, were measured with a stop-watch. The attachment positions to each FAST area were measured from the reference lines in centimeters with a ruler. The purpose of the FAST experiment was to determine whether the system performs sufficiently well for FAST. The evaluation indicators were set as follows: the robot continues extracting echo images during a patient’s body motion (rough breathing and coughing fits), the robot can extract echo images for FAST (pericardium, liver, spleen, and bladder) under a medical doctor’s control (position and orientation control), and the time required for FAST in each area. In this experiment, an examinee (BMI: 22) to whom the system was to be attached was separated from the physician in another room, and a real-time web camera and echo images and robot-control signals were transmitted between the examinee and the physician via a wired local network as shown in Fig. 3. For each FAST area, the time required for FAST, starting when the robot was attached to the patient and ending when the physician finished extracting the required echo image, was measured with a stopwatch. The target echo images extracted by the medical doctor were the images required for FAST: pericardium, liver, spleen, and bladder. The echo images obtained when patient body motion occurred were evaluated by a medical doctor. The number of trials in this FAST experiment was 5.
the overweight examinee was 256 ± 63 s (range 193–319 s). The robot could thus be attached to each FAST area of 3 body types in an average of approximately 3 min by a person with no medical knowledge. The attachment time to all 4 FAST areas of the overweight examinee was significantly longer than that of the slender examinee (p < 0.01). Additionally, the attachment time to all 4 FAST areas of the overweight examinee was significantly longer than that of the normal height and weight examinee (p < 0.05). The maximum gaps between the attaching positions by the examinees and the reference lines are shown in Tables 1 and 2. The maximum gap of the attachment positions in all 4 FAST areas to the slender examinee was 2.5 cm. The maximum gap of the attachment positions in all 4 FAST areas to the normal height and build examinee was 4.8 cm. The maximum gap of the attachment positions in all 4 FAST areas to the overweight examinee was 3.7 cm. The maximum gap was 2.7 cm for FAST area 1, 4.3 cm for FAST area 2, 4.8 cm for FAST area 3, and 2.5 cm for FAST area 4. Thus, the gaps for the
2.3. Statistical analysis A statistical analysis was performed using an analysis of variance (ANOVA) followed by Tukey’s post hoc test for multiple comparison. Statistical significance was defined as a p value <0.01, 0.05. 3. Results 3.1. Attachment experiment The results of the attachment experiment are shown in Fig. 4. There were no cases in which it was not possible to attach the robot. The attachment time to all 4 FAST areas of the slender examinee was 193 ± 58 s (range 135–251 s), the attachment time to all 4 FAST areas of the normal height and weight examinee was 216 ± 54 s (range 162–270 s), and the attachment time to all 4 FAST areas of
Fig. 4. Attachment time to each body type (N = 9). Although the person who attached the robot was a non-medical staff member, the robot could be attached to all FAST (Focused Assessment with Sonography for Trauma) areas within approximately 5 min (*p < 0.05, **p < 0.01).
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
169
Fig. 5. Echo images extracted by doctor’s remote control. Echo images could be extracted by the doctor’s remote control using the system. The echo images did not deteriorate during body motion. The doctor evaluated the echo images extracted from each FAST (Focused Assessment with Sonography for Trauma) area as suitable for FAST.
attachment positions in FAST areas 2 and 3 were larger than those for FAST areas 1 and 4. 3.2. FAST experiment by remote control Because the echo images and control signals were transmitted via a wired local network, the frame rate was 25–30 fps, and the communication delay was less than 30 ms. There was no Table 1 Maximum gap of the attachment positions in FAST areas 1 and 4. FAST 1
Slender Normal height and build Overweight
deterioration of the echo images when body motions, such as those due to rough breathing and coughing fits, occurred. The US probe did not come off the skin, and the echo images could be continuously extracted by the doctor using the remote-controlled the system on each FAST area. The doctor evaluated the extracted echo images by remote control on all the FAST areas (Fig. 5). The time required to obtain images from each FAST area are shown in Fig. 6. The time required to obtain images from FAST area 1 and FAST area 3 was 137 ± 97 s (range 40–234 s) and 136 ± 76 s (range 60–212 s), and these times were longer than the times required for FAST areas 2 and 4. There was also a great deal of
FAST 4
Left (cm)
Right (cm)
Left (cm)
Right (cm)
1.2 0.7
1.8 2.7
1.2 2.0
2.5 1.4
2
0.6
2.0
1.5
FAST, Focused Assessment with Sonography for Trauma; slender, body mass index (BMI) < 19; normal height and build, 19 ≤ BMI < 26; overweight, 26 ≤ BMI.
Table 2 Maximum gap of the attachment positions in FAST areas 2 and 3. FAST 2
Slender Normal height and build Overweight
FAST 3
Head (cm)
Foot (cm)
Head (cm)
Foot (cm)
1.7 1.4
1.4 4.3
0.6 1.9
0.6 4.8
0
3.0
0
3.7
FAST, Focused Assessment with Sonography for Trauma; slender, body mass index (BMI) < 19; normal height and build, 19 ≤ BMI < 26; overweight, 26 ≤ BMI.
Fig. 6. FAST (Focused Assessment with Sonography for Trauma) time on each area (N = 5). The time required to obtain FAST images from all the areas was approximately 354 ± 182 (range 172–536 s). The doctor could complete the FAST within 9 min (*p < 0.05).
170
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
Fig. 7. Comparison of time required for ambulance transportation and the proposed system. FAST (Focused Assessment with Sonography for Trauma) could be finished earlier with the proposed system.
variation in the times required for FAST areas 1 and 3. The time required for FAST area 4 was the shortest (15 ± 4 s, range 11–19 s). The time required for FAST area 1 was significantly longer than that of FAST area 4 (p < 0.05). Additionally, the time required for FAST area 3 was significantly longer than that of FAST area 4 (p < 0.05). 4. Discussion 4.1. Portability The previously developed tele-echography robotic systems are very large. The bed space in an ambulance is narrow, and this area is further minimized when these large systems are used. By reducing the available workspace in an ambulance, these systems interfere with other treatments and slow system setup. The proposed system is only 200 mm × 200 mm × 100 mm and can be placed, along with the controller and battery, into a in the suitcase measuring only 27 cm × 50 cm × 70 cm. The robot weighs only 2.2 kg. In addition, FAST can be started by simply attaching the robot to the patient, the system does not occupy bedside space, and the system can be set up quickly. As a result, the system is much more portable than previous systems. 4.2. Usability It was confirmed that different body types required significantly different attachment times. The analysis showed that larger patients required longer times to attach the system. It is possible that it was more difficult to wrap the corset around the back of a larger patient. For all the body types, it took the longest amount of time to attach the robot to FAST area 4. It is also possible that it was difficult to lift the buttocks of examinees to wrap the corset for the assessment of FAST area 4. The robot could, however, be attached to all the FAST areas within approximately 5 min, even by a non-medical staff member. Although the attachment positions in FAST areas 2 and 3 had larger gaps than the attachment positions in FAST areas 1 and 4, the maximum gaps of all the attachment positions on the FAST areas was 4.8 cm and could be compensated by remote-controlling the US probe. The robot could be attached to all the FAST areas regardless of body type. The use of landmarks, such as the nipple–nipple line and the umbilicus–sternum line, helped individuals to effectively attach the system. This suggests that a paramedic, who has received training in emergency medical care,
should be able to attach the robot to a patient even more quickly and accurately than one who has not.
4.3. Performance The doctor was able to perform FAST using the robot on each FAST area even during body motions such as rough breathing and coughing fits. Attaching the robot directly to the body trunk made it easier to adapt the patient body motion than using a control system. In addition, the echo images required for FAST could be obtained because the robot could control the US probe for FAST and because contact between the probe and the patient’s body was consistently maintained by the mechanical springs, which also corrected for irregularities in the body surface, such as a patient’s rib. The robot performance is considered good enough to obtain an echo image required for FAST. In the current study, the echo images were transmitted using a wired local network. An echo image transmission experiment will be conducted in an ambulance to examine how echo images are extracted by the system when the data are transferred via a mobile network. It was confirmed that the differences among the time required to obtain images from each of the FAST areas were significant (p < 0.01). Additionally, using multiple comparisons for the times required to obtain images from each FAST area, it was confirmed that the time required for FAST area 1 was significantly different than that of FAST area 4 (p < 0.05), and the time required for FAST area 3 was significantly different than that of FAST area 4 (p < 0.05). It is hypothesized that this is due to difficulty in quickly identifying the strenuous pericardium in FAST area 1. Further, in FAST area 3, it was hard to perceive the positional relationship between the spleen and the other organs. The time required to obtain images from FAST area 4 was shorter than that of the other FAST areas because the bladder can be extracted easily by echo, which is excellent at specifying liquid, enabling the doctor to easily extract the echo image. The current study showed that the system performs well enough for FAST to be completed within 9 min. To evaluate the effectiveness of the time required to attach the robot and perform FAST by remote control, this time was compared with the current ambulance transportation time in Japan (Fig. 7). The attachment time and the time required for FAST by remote control shown in Fig. 7 were based on the results of the maximum attachment time and maximum length of the FAST experiment, which were 5 min and 9 min, respectively (Figs. 4 and 6). It was
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
also assumed that the system was used quickly when the ambulance arrived at an injury site. It was confirmed that FAST would be completed by the end of transportation and that the remaining time would be approximately 11 min. Thus, this system could reduce the time needed to complete FAST, and it would be possible to treat a shock patient during the remaining time and to decide to which hospital to take the patient. This will make the proposed system effective in emergency medical care. We intend to further reduce the time needed for FAST and to make a system suitable for practical use in an ambulance. An experiment will need to be conducted to examine whether FAST could be conducted by a doctor using seamless wireless technology to remotely control the robot in an ambulance traveling on public roads [16,17], and a new attachment method will have to be developed so that the robot can be quickly attached to overweight patients. 5. Conclusion FAST is a quick and easy echography method for shock patients. The wearable tele-echography robot was developed and evaluated in the current study, and it was determined that this system is portable and easy to use for FAST in emergency situations. The proposed system was evaluated under a wired local network on male patients. Therefore, the effectiveness of the system is limited to these conditions. An enhanced version of the system that can be used with a wireless network will evaluate its practical use. In addition, a new attachment method for overweight and female patients will be considered in future improvements of this system. Role of the funding source This research was supported in part by the Ministry of Education, Culture, Sports, Science and Technology (Grants-in-Aid for Scientific Research, A07113600). The authors declare that the study sponsor had no role of the study design, the collection, analysis and interpretation of data; the writing of the manuscript; and the decision to submit the manuscript for publication. Ethical approval In this study, 13 healthy volunteers (19–29 years old) without any medical knowledge were included. The study was approved by the local research ethics committee (Approval number 2009-168, vol.560, Waseda University), and informed consent was obtained from all participants. An attachment experiment and a remotecontrolled FAST experiment were performed at Waseda University with medical doctors. A cushion was used to support the subjects’ body which were lifted by about 2 cm to avoid compressing their back against the bed.
171
Acknowledgements This research was supported in part by the Waseda Institute for Advanced Study (WIAS), the Ministry of Education, Culture, Sports, Science and Technology (Grants-in-Aid for Scientific Research, A07113600), and the Critical Care and Emergency Center at Yokohama City University Medical Center. Conflict of interest The authors declare that they have no competing interests. References [1] Mohamed MR, Fikri MA. Focussed assessment sonography trauma (FAST) and CT scan in blunt abdominal trauma: surgeon’s perspective. Afr Health Sci 2006;6:187–90. [2] Hoff WS, Holevar M, Nagy KK, Patterson L, Young JS, Arrillaga A, et al. Practice management guidelines for the evaluation of blunt abdominal trauma: the east practice management guidelines work group. J Trauma 2002;3: 602–15. [3] Rozycki GS, Pennington SD, Feliciano DV. Surgeon-performed ultrasound in trauma and surgical critical care setting: its use as an extension of the physical examination to detect pleural effusion trauma. J Trauma Injury Infect Crit Care 2001;4:636–42. [4] Huagui L, Marc W, Arthur E, William B, John W. Potential risk of vasovagal syncope for motor vehicle driving. Am Heart J 2000;2:184–6. [5] Phiroze H, Steven KB. The effect of epilepsy or diabetes mellitus on the risk of automobile accidents. N Engl J Med 1991;324:22–6. [6] Morrison JE, Wisner DH, Ramos LP. Syncope-related trauma: rationale and yield of diagnostic studies. J Trauma Injury Infect Crit Care 1999;4:707–10. [7] Esposito TJ, Sanddal ND, Hansen JD, Reynolds S. Analysis of preventable trauma deaths and inappropriate trauma care in a rural state. J Trauma Injury Infect Crit Care 1998;5:955–62. [8] Zhu WH, Salcudean SE, Bachmann S, Abolmaesumi P. Motion/force/image control of a diagnostic ultrasound robot. Proc IEEE Int Conf Robot Autom 2000;2:1580–5. [9] Adriana V, Jocelyne T, Philippe C, Agnes G, Franck P, Pierre T, et al. Experiments with the TER tele-echography robot. Proc IEEE Med Image Comput Comput Assist Interv 2002;2488:138–46. [10] Vilchis A, Troccaz T, Cinquin P, Masuda K, Pellissier F. A new robot architecture for tele-echography. IEEE Trans Robot Autom 2003;5:922–6. [11] Masuda K, Horiguchi T, Okamori K, Watanabe H, Ozawa K, Yoshinaga T, et al. Development of a twist pantograph mechanism for robotic tele-echography. Proc 4th Eur Conf Int Fed Med Biol Eng 2005;8:927–31. [12] Masuda K, Tateishi N, Suzuki Y, Kimura E, Wie Y, Ishihara K. Experiment of wireless tele-echography system by controlling echographic diagnosis robot. Proc Med Image Comput Comput Assist Interv 2002;2488:130–7. [13] Takeuchi R, Harada H, Masuda K, Ota G, Yokoi M, Teramura N, et al. Field testing of a remote controlled robotic tele-echo system in an ambulance using broadband mobile communication technology. J Med Syst 2008;3: 235–42. [14] Robosoft Inc. Estele. Advanced Robotics Solutions; 2007, http://robosoft.com. [15] Koizumi N, Warisawa N, Nagoshi M, Hashizume H, Mitsuishi M. A study on the construction methodology of remote ultrasound diagnosis system. J Robot Soc 2007;2:267–79. [16] Abolmaesumi P, Salcudean SE, Zhu WH, Sirouspour M, DiMaio SP. Image-guided control of a robot for medical ultrasound. IEEE Trans Robot 2002;1:11–23. [17] Christofer AS, Bernard JR, Robert TG, Frank LC, James RB, Terry DB, et al. Satellite and mobile wireless transmission of focused assessment with sonography in trauma. Acad Emerg Med 2003;12:1411–4.
Contents lists available at SciVerse ScienceDirect
Medical Engineering & Physics journal homepage: www.elsevier.com/locate/medengphy
Usability and performance of a wearable tele-echography robot for focused assessment of trauma using sonography Keiichiro Ito a,∗ , Shigeki Sugano a , Ryohei Takeuchi b , Kyota Nakamura c , Hiroyasu Iwata d a
Department of Creative Science and Engineering, Waseda University, Japan Department of Joint Surgery Center, Yokosuka Municipal Hospital, Japan Department of Emergency Medicine, Yokohama City University Hospital, Japan d Waseda Institute for Advanced Study (WIAS), Waseda University, Japan b c
a r t i c l e
i n f o
Article history: Received 6 January 2012 Received in revised form 27 April 2012 Accepted 28 April 2012 Keywords: Emergency medicine Tele-echography Wearable system
a b s t r a c t Focused Assessment with Sonography for Trauma (FAST) is widely used as a first lifesaving step for patients suffering from internal bleeding. Because it may take a long time to transport such patients to a hospital, a wearable and portable tele-echography robot that a paramedic can attach to the patient has been developed. In the current study, experiments were conducted to evaluate the usability and performance of attached FAST. The proposed robot must be attached to 4 areas to perform FAST. The time required for attachment and the positions of attachment completed by 9 non-medical staff members, as well as the time it took for the FAST to reach a medical doctor, were measured. The echo images obtained when the patient’s body was in motion were evaluated by a medical doctor. The robot could be attached to all 4 areas within approximately 5 min, and the maximum gap was 4.8 cm. This indicates that a paramedic who has received training in emergency medical care should be able to attach the robot to a patient quickly and accurately. Additionally, it was confirmed that the robot could be used to complete FAST under a doctor’s control within 9 min and that the extracted echo images were suitable for FAST. A comparison of the results with current ambulance transportation time confirmed that FAST could be completed approximately 14 min before the patient reached the hospital. The results of the current study indicate that the robot is worth using, is suitable for FAST, and will be effective in emergency medical care. © 2012 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Echography is a critical diagnostic approach in the treatment of shock patients who are severely injured in traffic accidents or large-scale disasters, and Focused Assessment with Sonography for Trauma (FAST) is a quick, easy, and effective echography method that is widely used in the initial hospital evaluation of traumatic shock patients [1–7]. Because transporting a patient from an injury site to a hospital often takes a long time, a tele-echography system that a paramedic can easily use at the injury site or in the ambulance should increase the patient’s likelihood of surviving. The system’s portability and ease of use by paramedics are very important. The system should be small enough to be easily used in an ambulance,
∗ Corresponding author at: Department of Creative Science and Engineering, School of Modern Mechanical Engineering, Waseda University, 17 Kikui-cho, Shinjuku-ku, Tokyo 162-0044, Japan. Tel.: +81 3 3203 4382; fax: +81 3 3203 4382. E-mail address: [email protected] (K. Ito).
and it should be easy to handle, allowing a paramedic to use it quickly and properly. Many tele-echography systems have been developed recently. Zhu et al., for example, developed a robotic system for carotid artery tele-echography [8]. Vilchis et al. developed a tele-echography system in which a physician adjusts the location of an echo probe moved by a cable-controlled robot strapped to the body of a patient at a distant site [9,10]. Masuda et al. developed a wirelessly controlled system with an ultrasound (US) probe mounted on the arm of a pantograph mechanism attached to the side of the patient’s bed [11,12]. Finally, through the field testing of a mobile robotic teleecho system placed in an ambulance, Takeuchi et al. have confirmed that this system could be remotely operated and send clear, realtime US images of a patient’s abdomen to the destination hospital [13]. These systems, however, are hard to carry and require a lot of bedside space. Robosoft Inc. therefore developed a portable teleechography robotic system [14] that is pushed against the patient’s body by a medical assistant while a physician remotely controls the position of the US probe. Although the system is easy to transport,
1350-4533/$ – see front matter © 2012 IPEM. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.medengphy.2012.04.011
166
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
it is not easy for medical assistants to use because they have to determine the location on the patient’s body to apply pressure to activate the system by interpreting the often vague instructions of a doctor, and they must apply constant pressure. The purpose of the work presented here was to experimentally evaluate the usability and performance of a prototype wearable tele-echography robot designed to be used easily by a paramedic in an ambulance. 2. Materials and methods
site. Information concerning the position and orientation of the US probe, as well as image and voice information, is also transmitted. The doctor performs the tele-echography while observing the echo image, the US probe, and the patient. The portable echo device consists of a MicroMaxx (SonoSite Inc.) 30 cm wide, 27 cm long, and 8 cm high (weight: 3.9 kg) and a sector US probe (SonoSite Inc., P17/5-1). The frequency range of the US probe used for the chest and abdomen is 1–5 MHz. A portable battery for the system is also required. In addition, different types of networks are required for the system: ISDN (typically 128 kb/s), LAN, and WiMAX (IEEE 802.16e, download: 40 Mb/s, upload: 10 Mb/s).
2.1. Overview of the wearable tele-echography robot system 2.2. Wearable tele-echography robot An overview of the wearable tele-echography robot system is shown in Fig. 1(a). A medical doctor operates the graphical user interface on the computer, and the control signals are transmitted through a network to the robot in an ambulance or an injury
The robot can be easily attached to the trunk of the patient’s body using a corset belt, and it naturally adapts to the patient’s body motion. It has a control mechanism that enables it to perform
Fig. 1. Proposed wearable tele-echography robotic system. (a) System overview. To control the wearable tele-echography robot and transmit audio and camera data, a mobile network will be used for the connection between the hospital and the patient. The system fits into a suitcase 27 cm wide, 50 cm long, and 70 cm high. (b) Wearable teleechography robot. The robot provides 4 degrees of freedom (pitching, rolling, positioning, and contacting) and control of the US probe for FAST. Additionally, the robot has two curvature rails, a soft urethane sponge, an elastic silicon-based corset, two rotary motors, a linear motor, and two mechanical springs. It can be attached to a patient by wrapping the corset around the patient’s trunk. The probe is kept in contact with the patient’s skin by two mechanical springs. The robot measures 200 mm × 200 mm × 100 mm and weighs 2.2 kg.
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
echography, and it can generate a force that keeps the US probe in contact with the patient’s body. As shown in Fig. 1(b), the robot has 4 degrees of freedom: pitching, rolling, positioning, and contacting. The curvature rails and rotary motors (Harmonic drive, RSF-5A, 66 g) are used to produce the pitching and rolling motions of the probe: 90 degrees pitching and 360 degrees rolling. A linear motor (Hitachi, S080Q, weight; 80 g) is used to position the probe and is able move it as much as 100 mm. The urethane sponge is used as a cushion between the robot and the patient’s skin, and the two mechanical springs (Samini Inc., Compression spring, Coefficient Kf; 0.27 N/mm) push the US probe against the patient. In designing the probe control mechanism for echography, the main design objectives were to make the mechanism simple and to make the robot small. Because the US probe needs to be in contact with the body surface even during pitching, rolling, and motion across the body surface, a mechanism that generates contact force passively with two mechanical springs was developed. The contact force is generated by attaching the robot; the probe is pushed onto the surface of the body. Additionally, the proposed mechanism keeps the point of the probe in contact with the patient’s skin, and a contact force is continuously generated when the probe pitches, rolls, and moves across the body from springs installed in the end of the robot. Koizumi et al. reported that the contact force of the US probe should be no more than 9 N so that the area of diagnosis is not pressed
167
excessively by the probe and the echo image is not distorted [15]. The stroke of the two springs was therefore set to 10 mm. When the robot is attached to the patient, this mechanism can generate a contact force of approximately 5.4 N. The usability and performance of the system for FAST were evaluated in 2 experiments: an attachment experiment using examinees and a remote-controlled FAST experiment. Because FAST narrows the area of diagnosis down to 4 parts—pericardium, liver, spleen, and bladder—the attachment time and position in all 4 areas were used as indicators of the usability of the system. In this experiment, examinees were divided into 2 groups: examinees attaching the robot and examinees having the robot attached to them. The patient’s body type is a key indicator for how the robot should be attached. The examinees having the robot attached to them were classified into 3 types (slender, normal, and overweight) based on their BMIs (BMI: 18, 22, and 28, respectively). Because the skill level of actual paramedics would potentially affect the result of the experiment, the minimum usability of the system was evaluated using 9 non-medical staff members who had the same low level of skill. The examinees attached the robot to all 4 areas while looking at a figure of attachment positions and procedures (Fig. 2). Attaching the reference lines was based on the line through the umbilicus and the middle of the sternum and the line through the nipples. The attachment times, starting when the robot was picked up and ending when the robot had been attached to each
Fig. 2. Attachment position and attaching procedure. The examinees attached the robot on each body type, without instructions from the doctor, by looking at the attachment position and the pictures showing the procedure. The reference lines for attachment were based on the line between the nipples (for FAST (Focused Assessment with Sonography for Trauma) areas 1, 2, and 3) and the line between the umbilicus and the center of the sternum (for FAST area 4); thus, even non-medical personnel could easily understand the attachment positions. The examinees attached the robot along the reference lines. The best attachment positions of the robot were assumed to be the areas where FAST can be performed by the doctor’s remote control. The measurement of the gaps between the attachment positions and FAST areas 1 and 4 were based on the umbilicus–sternum line, and the measurement of the gaps between the attachment positions and FAST areas 2 and 3 were based on nipple–nipple line.
168
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
Fig. 3. FAST (Focused Assessment with Sonography for Trauma) experiment. The robot was attached to FAST 1 area based on the result of the attachment experiment, which was defined as the maximum gap. The doctor remotely controlled the robot via a wired local network, and the time required to extract the echo image for FAST area 1 (pericardium) was measured. The control of the US (ultrasound) probe and the extraction of the echo image were conducted in the same way on each FAST area, and times required to obtain FAST images for each area were also measured. The doctor evaluated the sharpness of the echo images obtained for each FAST area with regard to their suitability for FAST.
FAST area, were measured with a stop-watch. The attachment positions to each FAST area were measured from the reference lines in centimeters with a ruler. The purpose of the FAST experiment was to determine whether the system performs sufficiently well for FAST. The evaluation indicators were set as follows: the robot continues extracting echo images during a patient’s body motion (rough breathing and coughing fits), the robot can extract echo images for FAST (pericardium, liver, spleen, and bladder) under a medical doctor’s control (position and orientation control), and the time required for FAST in each area. In this experiment, an examinee (BMI: 22) to whom the system was to be attached was separated from the physician in another room, and a real-time web camera and echo images and robot-control signals were transmitted between the examinee and the physician via a wired local network as shown in Fig. 3. For each FAST area, the time required for FAST, starting when the robot was attached to the patient and ending when the physician finished extracting the required echo image, was measured with a stopwatch. The target echo images extracted by the medical doctor were the images required for FAST: pericardium, liver, spleen, and bladder. The echo images obtained when patient body motion occurred were evaluated by a medical doctor. The number of trials in this FAST experiment was 5.
the overweight examinee was 256 ± 63 s (range 193–319 s). The robot could thus be attached to each FAST area of 3 body types in an average of approximately 3 min by a person with no medical knowledge. The attachment time to all 4 FAST areas of the overweight examinee was significantly longer than that of the slender examinee (p < 0.01). Additionally, the attachment time to all 4 FAST areas of the overweight examinee was significantly longer than that of the normal height and weight examinee (p < 0.05). The maximum gaps between the attaching positions by the examinees and the reference lines are shown in Tables 1 and 2. The maximum gap of the attachment positions in all 4 FAST areas to the slender examinee was 2.5 cm. The maximum gap of the attachment positions in all 4 FAST areas to the normal height and build examinee was 4.8 cm. The maximum gap of the attachment positions in all 4 FAST areas to the overweight examinee was 3.7 cm. The maximum gap was 2.7 cm for FAST area 1, 4.3 cm for FAST area 2, 4.8 cm for FAST area 3, and 2.5 cm for FAST area 4. Thus, the gaps for the
2.3. Statistical analysis A statistical analysis was performed using an analysis of variance (ANOVA) followed by Tukey’s post hoc test for multiple comparison. Statistical significance was defined as a p value <0.01, 0.05. 3. Results 3.1. Attachment experiment The results of the attachment experiment are shown in Fig. 4. There were no cases in which it was not possible to attach the robot. The attachment time to all 4 FAST areas of the slender examinee was 193 ± 58 s (range 135–251 s), the attachment time to all 4 FAST areas of the normal height and weight examinee was 216 ± 54 s (range 162–270 s), and the attachment time to all 4 FAST areas of
Fig. 4. Attachment time to each body type (N = 9). Although the person who attached the robot was a non-medical staff member, the robot could be attached to all FAST (Focused Assessment with Sonography for Trauma) areas within approximately 5 min (*p < 0.05, **p < 0.01).
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
169
Fig. 5. Echo images extracted by doctor’s remote control. Echo images could be extracted by the doctor’s remote control using the system. The echo images did not deteriorate during body motion. The doctor evaluated the echo images extracted from each FAST (Focused Assessment with Sonography for Trauma) area as suitable for FAST.
attachment positions in FAST areas 2 and 3 were larger than those for FAST areas 1 and 4. 3.2. FAST experiment by remote control Because the echo images and control signals were transmitted via a wired local network, the frame rate was 25–30 fps, and the communication delay was less than 30 ms. There was no Table 1 Maximum gap of the attachment positions in FAST areas 1 and 4. FAST 1
Slender Normal height and build Overweight
deterioration of the echo images when body motions, such as those due to rough breathing and coughing fits, occurred. The US probe did not come off the skin, and the echo images could be continuously extracted by the doctor using the remote-controlled the system on each FAST area. The doctor evaluated the extracted echo images by remote control on all the FAST areas (Fig. 5). The time required to obtain images from each FAST area are shown in Fig. 6. The time required to obtain images from FAST area 1 and FAST area 3 was 137 ± 97 s (range 40–234 s) and 136 ± 76 s (range 60–212 s), and these times were longer than the times required for FAST areas 2 and 4. There was also a great deal of
FAST 4
Left (cm)
Right (cm)
Left (cm)
Right (cm)
1.2 0.7
1.8 2.7
1.2 2.0
2.5 1.4
2
0.6
2.0
1.5
FAST, Focused Assessment with Sonography for Trauma; slender, body mass index (BMI) < 19; normal height and build, 19 ≤ BMI < 26; overweight, 26 ≤ BMI.
Table 2 Maximum gap of the attachment positions in FAST areas 2 and 3. FAST 2
Slender Normal height and build Overweight
FAST 3
Head (cm)
Foot (cm)
Head (cm)
Foot (cm)
1.7 1.4
1.4 4.3
0.6 1.9
0.6 4.8
0
3.0
0
3.7
FAST, Focused Assessment with Sonography for Trauma; slender, body mass index (BMI) < 19; normal height and build, 19 ≤ BMI < 26; overweight, 26 ≤ BMI.
Fig. 6. FAST (Focused Assessment with Sonography for Trauma) time on each area (N = 5). The time required to obtain FAST images from all the areas was approximately 354 ± 182 (range 172–536 s). The doctor could complete the FAST within 9 min (*p < 0.05).
170
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
Fig. 7. Comparison of time required for ambulance transportation and the proposed system. FAST (Focused Assessment with Sonography for Trauma) could be finished earlier with the proposed system.
variation in the times required for FAST areas 1 and 3. The time required for FAST area 4 was the shortest (15 ± 4 s, range 11–19 s). The time required for FAST area 1 was significantly longer than that of FAST area 4 (p < 0.05). Additionally, the time required for FAST area 3 was significantly longer than that of FAST area 4 (p < 0.05). 4. Discussion 4.1. Portability The previously developed tele-echography robotic systems are very large. The bed space in an ambulance is narrow, and this area is further minimized when these large systems are used. By reducing the available workspace in an ambulance, these systems interfere with other treatments and slow system setup. The proposed system is only 200 mm × 200 mm × 100 mm and can be placed, along with the controller and battery, into a in the suitcase measuring only 27 cm × 50 cm × 70 cm. The robot weighs only 2.2 kg. In addition, FAST can be started by simply attaching the robot to the patient, the system does not occupy bedside space, and the system can be set up quickly. As a result, the system is much more portable than previous systems. 4.2. Usability It was confirmed that different body types required significantly different attachment times. The analysis showed that larger patients required longer times to attach the system. It is possible that it was more difficult to wrap the corset around the back of a larger patient. For all the body types, it took the longest amount of time to attach the robot to FAST area 4. It is also possible that it was difficult to lift the buttocks of examinees to wrap the corset for the assessment of FAST area 4. The robot could, however, be attached to all the FAST areas within approximately 5 min, even by a non-medical staff member. Although the attachment positions in FAST areas 2 and 3 had larger gaps than the attachment positions in FAST areas 1 and 4, the maximum gaps of all the attachment positions on the FAST areas was 4.8 cm and could be compensated by remote-controlling the US probe. The robot could be attached to all the FAST areas regardless of body type. The use of landmarks, such as the nipple–nipple line and the umbilicus–sternum line, helped individuals to effectively attach the system. This suggests that a paramedic, who has received training in emergency medical care,
should be able to attach the robot to a patient even more quickly and accurately than one who has not.
4.3. Performance The doctor was able to perform FAST using the robot on each FAST area even during body motions such as rough breathing and coughing fits. Attaching the robot directly to the body trunk made it easier to adapt the patient body motion than using a control system. In addition, the echo images required for FAST could be obtained because the robot could control the US probe for FAST and because contact between the probe and the patient’s body was consistently maintained by the mechanical springs, which also corrected for irregularities in the body surface, such as a patient’s rib. The robot performance is considered good enough to obtain an echo image required for FAST. In the current study, the echo images were transmitted using a wired local network. An echo image transmission experiment will be conducted in an ambulance to examine how echo images are extracted by the system when the data are transferred via a mobile network. It was confirmed that the differences among the time required to obtain images from each of the FAST areas were significant (p < 0.01). Additionally, using multiple comparisons for the times required to obtain images from each FAST area, it was confirmed that the time required for FAST area 1 was significantly different than that of FAST area 4 (p < 0.05), and the time required for FAST area 3 was significantly different than that of FAST area 4 (p < 0.05). It is hypothesized that this is due to difficulty in quickly identifying the strenuous pericardium in FAST area 1. Further, in FAST area 3, it was hard to perceive the positional relationship between the spleen and the other organs. The time required to obtain images from FAST area 4 was shorter than that of the other FAST areas because the bladder can be extracted easily by echo, which is excellent at specifying liquid, enabling the doctor to easily extract the echo image. The current study showed that the system performs well enough for FAST to be completed within 9 min. To evaluate the effectiveness of the time required to attach the robot and perform FAST by remote control, this time was compared with the current ambulance transportation time in Japan (Fig. 7). The attachment time and the time required for FAST by remote control shown in Fig. 7 were based on the results of the maximum attachment time and maximum length of the FAST experiment, which were 5 min and 9 min, respectively (Figs. 4 and 6). It was
K. Ito et al. / Medical Engineering & Physics 35 (2013) 165–171
also assumed that the system was used quickly when the ambulance arrived at an injury site. It was confirmed that FAST would be completed by the end of transportation and that the remaining time would be approximately 11 min. Thus, this system could reduce the time needed to complete FAST, and it would be possible to treat a shock patient during the remaining time and to decide to which hospital to take the patient. This will make the proposed system effective in emergency medical care. We intend to further reduce the time needed for FAST and to make a system suitable for practical use in an ambulance. An experiment will need to be conducted to examine whether FAST could be conducted by a doctor using seamless wireless technology to remotely control the robot in an ambulance traveling on public roads [16,17], and a new attachment method will have to be developed so that the robot can be quickly attached to overweight patients. 5. Conclusion FAST is a quick and easy echography method for shock patients. The wearable tele-echography robot was developed and evaluated in the current study, and it was determined that this system is portable and easy to use for FAST in emergency situations. The proposed system was evaluated under a wired local network on male patients. Therefore, the effectiveness of the system is limited to these conditions. An enhanced version of the system that can be used with a wireless network will evaluate its practical use. In addition, a new attachment method for overweight and female patients will be considered in future improvements of this system. Role of the funding source This research was supported in part by the Ministry of Education, Culture, Sports, Science and Technology (Grants-in-Aid for Scientific Research, A07113600). The authors declare that the study sponsor had no role of the study design, the collection, analysis and interpretation of data; the writing of the manuscript; and the decision to submit the manuscript for publication. Ethical approval In this study, 13 healthy volunteers (19–29 years old) without any medical knowledge were included. The study was approved by the local research ethics committee (Approval number 2009-168, vol.560, Waseda University), and informed consent was obtained from all participants. An attachment experiment and a remotecontrolled FAST experiment were performed at Waseda University with medical doctors. A cushion was used to support the subjects’ body which were lifted by about 2 cm to avoid compressing their back against the bed.
171
Acknowledgements This research was supported in part by the Waseda Institute for Advanced Study (WIAS), the Ministry of Education, Culture, Sports, Science and Technology (Grants-in-Aid for Scientific Research, A07113600), and the Critical Care and Emergency Center at Yokohama City University Medical Center. Conflict of interest The authors declare that they have no competing interests. References [1] Mohamed MR, Fikri MA. Focussed assessment sonography trauma (FAST) and CT scan in blunt abdominal trauma: surgeon’s perspective. Afr Health Sci 2006;6:187–90. [2] Hoff WS, Holevar M, Nagy KK, Patterson L, Young JS, Arrillaga A, et al. Practice management guidelines for the evaluation of blunt abdominal trauma: the east practice management guidelines work group. J Trauma 2002;3: 602–15. [3] Rozycki GS, Pennington SD, Feliciano DV. Surgeon-performed ultrasound in trauma and surgical critical care setting: its use as an extension of the physical examination to detect pleural effusion trauma. J Trauma Injury Infect Crit Care 2001;4:636–42. [4] Huagui L, Marc W, Arthur E, William B, John W. Potential risk of vasovagal syncope for motor vehicle driving. Am Heart J 2000;2:184–6. [5] Phiroze H, Steven KB. The effect of epilepsy or diabetes mellitus on the risk of automobile accidents. N Engl J Med 1991;324:22–6. [6] Morrison JE, Wisner DH, Ramos LP. Syncope-related trauma: rationale and yield of diagnostic studies. J Trauma Injury Infect Crit Care 1999;4:707–10. [7] Esposito TJ, Sanddal ND, Hansen JD, Reynolds S. Analysis of preventable trauma deaths and inappropriate trauma care in a rural state. J Trauma Injury Infect Crit Care 1998;5:955–62. [8] Zhu WH, Salcudean SE, Bachmann S, Abolmaesumi P. Motion/force/image control of a diagnostic ultrasound robot. Proc IEEE Int Conf Robot Autom 2000;2:1580–5. [9] Adriana V, Jocelyne T, Philippe C, Agnes G, Franck P, Pierre T, et al. Experiments with the TER tele-echography robot. Proc IEEE Med Image Comput Comput Assist Interv 2002;2488:138–46. [10] Vilchis A, Troccaz T, Cinquin P, Masuda K, Pellissier F. A new robot architecture for tele-echography. IEEE Trans Robot Autom 2003;5:922–6. [11] Masuda K, Horiguchi T, Okamori K, Watanabe H, Ozawa K, Yoshinaga T, et al. Development of a twist pantograph mechanism for robotic tele-echography. Proc 4th Eur Conf Int Fed Med Biol Eng 2005;8:927–31. [12] Masuda K, Tateishi N, Suzuki Y, Kimura E, Wie Y, Ishihara K. Experiment of wireless tele-echography system by controlling echographic diagnosis robot. Proc Med Image Comput Comput Assist Interv 2002;2488:130–7. [13] Takeuchi R, Harada H, Masuda K, Ota G, Yokoi M, Teramura N, et al. Field testing of a remote controlled robotic tele-echo system in an ambulance using broadband mobile communication technology. J Med Syst 2008;3: 235–42. [14] Robosoft Inc. Estele. Advanced Robotics Solutions; 2007, http://robosoft.com. [15] Koizumi N, Warisawa N, Nagoshi M, Hashizume H, Mitsuishi M. A study on the construction methodology of remote ultrasound diagnosis system. J Robot Soc 2007;2:267–79. [16] Abolmaesumi P, Salcudean SE, Zhu WH, Sirouspour M, DiMaio SP. Image-guided control of a robot for medical ultrasound. IEEE Trans Robot 2002;1:11–23. [17] Christofer AS, Bernard JR, Robert TG, Frank LC, James RB, Terry DB, et al. Satellite and mobile wireless transmission of focused assessment with sonography in trauma. Acad Emerg Med 2003;12:1411–4.