American Journal of Emergency Medicine (2012) 30, 961-965

Clinical Notes

Medical consultations and the sharing of medical images involving spinal injury over mobile phone networks

Michal Filip MD, PhD ^a, Petr Linzer MD ^a, Filip Samal MD ^a, Jiri Tesar MD ^b,

Roman Herzig MD, PhD ^c, David Skoloudik MD, PhD ^c,d,?

^aDepartment of Neurosurgery, Bata Hospital, CZ-76001, Zlin, Czech Republic

^bDepartment of Radiology, Bata Hospital, CZ-76001, Zlin, Czech Republic

^cDepartment of Neurology, Palacky University Medical School and University Hospital, CZ-77520, Olomouc, Czech Republic

^dDepartment of Neurology, Ostrava University Medical School and University Hospital, CZ-70852, Ostrava, Czech Republic

Received 28 March 2011; revised 10 May 2011; accepted 11 May 2011

Abstract

Background: The transmission of medical images and other data over mobile phone networks may facilitate remote medical consultations between neurosurgeons and Regional hospitals treating spinal injury patients. The aim of this study was to compare the efficacy of mobile phone consultations with standard hospital workstation consultations in spinal injury patients.

Methods: The images were exported over the Internet from surrounding local hospitals through the Picture Archiving and Communication System, in DICOM III format, to the central hospital server. The xVision browser was used to view the acquired images on a standard workstation. The data were also exported to the secured hospital Web server IIS60 and converted to JPEG format to enable remote physician access and consultation. The remote consulting physician connected to this server by mobile phone using the phone’s Internet browser. A second physician, blind to the mobile phone results, evaluated the same images at a workstation in the hospital. The results of the mobile phone consultations were compared with the results from standard workstation consultations.

Results: There was no difference in the quality of spinal computed tomographic/magnetic resonance images viewed on the phone screen compared with on the workstation. More importantly, the final diagnoses made by mobile phone did not differ from those made by workstation consultations. A transfer to the department of neurosurgery was required after consultation in 11 patients.

Conclusion: Mobile phone consultations for patients with spinal injuries was as effective as workstation consultations. Mobile phone consultations can increase the expertise available to regional hospitals, which are often the First responders to medical emergencies.

(C) 2012

Introduction

* Corresponding author. Department of Neurology, Palacky University Medical School, CZ-77520 Olomouc, Czech Republic. Tel.: +42 0597375613.

E-mail address: [email protected] (D. Skoloudik).

Until recently, neurosurgeons have relied on telephone transmission or the physical transport of medical docu- ments and images for evaluation to remote sites, such as regional hospitals. Enhancing the speed of medical image transfer without sacrificing image resolution may improve

0735-6757/$ - see front matter (C) 2012 doi:10.1016/j.ajem.2011.05.007

remote care and clinical outcomes. This problem has been a major focus for health care information technologists worldwide. The first reports on building effective consultation networks appeared in the 1980s [1-11]. These networks used fixed connections and several types of data transfer protocols (Integrated Services Digital Network, Asymmetric Digital Subscriber Line) and user interfaces (e-mail, Web pages) [5,6,10-12]. The consulta- tion networks enable the evaluation of medical images from spine injury patients [1,4,7,12-14]. The use of a virtual private network (VPN) enables safe and private Internet access to the images. The attending neurosurgeon is able to view the images and can make an informed decision about further treatment, such as recommending transport to a specialized health care center or treatment at the local hospital [5,7,12,15]. The neurosurgeon can also perform consultations from home (when on call) using a home computer connected to the network by any type of broadband Internet connection that allows sharing of the workstation screen images. Whether the attending physi- cian can work at home or must be at the hospital depends on which location has the best fixed Internet connection [3,7,8,10,16].

High-quality home Internet connections allow physicians outside the hospital to participate in an increasing number of medical situations requiring remote consultation. The use of mobile phones could greatly increase the access of regional facilities to specialized expertise. The aim of this study was to compare the quality of remote consultations made over a mobile phone network to those made using standard hospital workstations in cases involving patients with spinal injuries admitted to regional hospitals.

Methods

The specialized services of the Bata Hospital (Zlin, Czech Republic) include consultations and the treatment of patients with spinal injuries admitted to 5 other regional hospitals. At this site, there is full-time neurosurgical service provided by a neurosurgeon or a resident in neurosurgery. When the resident on duty has fewer than 6 years of experience, an experienced neurosurgeon is available on call. Both in Bata Hospital and in the regional hospitals with emergency services for spinal injuries, a radiologist is present on a full-time basis. When a neuroSurgical consultation is needed, the radiologist exports computed tomography (CT) or magnetic resonance imaging (MRI) scans by means of Picture Archiving and Commu- nication System (PACS) to Bata Hospital. Consequently, the images are evaluated by the attending neurosurgeon and uploaded to a mobile phone. In the case of a neurosurgical resident on duty at Bata Hospital, a senior neurosurgeon is contacted; and he/she uploads the relevant data (typically 10 images) to a mobile phone.

Enrollment of the patients

Out of the 102 Neurosurgery consultations involving patients with diagnosed or suspected spinal injuries that were entered into the PACS from September 2009 to August 2010,

17 met our inclusion and exclusion criteria and were included in the prospective study. The inclusion criteria were as follows: (1) the patient’s age must be between 18 and 90 years; (2) a resident in neurosurgery must be on duty; and

(3) the patient must exhibit a spinal injury that is suspected to be unstable. The exclusion criterion was the presence of an on-duty experienced neurosurgeon. The neurosurgeons at Bata Hospital were consulted to verify a diagnosis and suggest further treatment options. The resident in neurosur- gery was on duty at a workstation, whereas the experienced neurosurgeon remotely used a mobile phone.

Workstation consultation

The images were exported, in DICOM III format, via the Internet using the PACS system from the surrounding hospitals to the central server of Bata Hospital in Zlin. Under normal conditions, the VPN network is able to transfer data (ie, medical images) with a download speed of 2 to 10 Mb/s 24 hours a day. Once downloaded to the central server, images and other documents were archived using the PACS system for future reference.

The neurosurgeon on duty evaluated the location and stability of the spinal injury. Later, the neurosurgeon and radiologist evaluated the quality of the CT and MRI scans on the workstation monitor. The time (in minutes) from exporting the images from the regional hospital to the final medical decision was recorded and evaluated.

Mobile phone consultation

After receipt of the message from the local hospital, the neurosurgeon on half-service/call informed the department staff and entered data on the patient being treated. The medical technician sought the relevant images using the xVision browser (in DICOM III format). By pressing the “export into HTML” button in the xVision browser, the set of images was sent, in JPEG format, to the secured Internet server IIS60. The consulting physician used a phone’s Internet browser to connect to the hospital’s Internet server using a preset user name and password. The connection speed was 512 kb/s. For this method of transfer, we used the PACS system, closed hospital Internet pages, an Internet connection using the mobile operator, and a mobile phone with a high-quality display (Nokia C6, Nokia, Espoo, Finland; display: TFT; 360 x 640; 16 million colors). As soon as the consulting physician was connected, JPEG images could be downloaded at a resolution of 200 x 200 pixels (?20 kb in size) to the mobile phone. After downloading data to the device, the physician evaluated

and selected the necessary images. The physician could zoom out to view the image at full resolution (512 x 512 pixels or ?60 kb for a typical CT scan) by clicking on it. After the evaluation of the images, the physician informed the local hospital by mobile phone of his or her recommendations for treatment.

The neurosurgeon evaluated the lesion location and whether it was a stable or unstable spinal injury. Later, the neurosurgeon and radiologist evaluated the CT and MRI scan quality on the mobile phone display.

All patients with unstable spinal injuries with or without cord lesions were transferred to the Department of Neurosurgery at Bata Hospital after agreement between the 2 consulting physicians (one on a mobile phone and one at a workstation). The time required to send images from the local hospital and complete the phone consultation was recorded.

Statistics

The paired Student t test was applied to assess the statistical significance of differences in the mean times required for mobile phone and workstation consultations. Statistical analyses were performed using SPSS software version 14.0 (IMB, SPSS Inc., Chicago, IL, USA).

Ethics approval

The study was conducted in accordance with the Helsinki Declaration of 1975 (as revised in 1983 and 2008), and it was approved by the local ethics committees of all hospitals involved. Each patient gave informed consent.

Results

Over this yearlong study, 17 patients with spinal injury that were admitted to a surrounding regional hospital met our inclusion and exclusion criteria and were included in the study. Demographic data are presented in Table 1. The causes of the spinal injuries were as follows: a fall from a height (8 patients), a car accident (5 patients), a sport-related

Table 1 Demographic data of spinal injury patients

No. of patients 17

Mean age +- SD (y) 48.2 +- 12.1

Male sex 62%

Upper cervical spinal injury 2 (11.8%)

Lower cervical spinal injury 4 (23.5%)

Thoracolumbar spinal injury 11 (64.7%)

Stable spinal injury 6 (35.3%)

Unstable spinal injury 11 (64.7%)

Injury-to-emergency department 89.2 +- 21.5 mean time +- SD (min)

Table 2 Time elapsed between sending images and the completion of consultation by workstation or mobile phone and Image quality

Workstation monitor

Mean time +- SD for 9.8 +- 2.2 sending images and consultation (min)

Mobile phone display

17.8 +- 2.4

P value

b.001

High quality of images (%)

17 (100%)

17 (100%)

N.05

injury (2 patients), or a work-related injury (2 patients). The injuries were located in the upper cervical spine (2 patients), the lower cervical spine (4 patients), or the thoracolumbar spine (11 patients). Six of these injuries were evaluated as stable and 11 as unstable.

The process of downloading and evaluating the images required significantly more time using the mobile phone compared with the workstation (Table 2), but it did not exceed 23 minutes for any patient. The neurosurgeon, in cooperation with the radiologist, determined that the quality of all CT and MRI scans was of sufficient resolution for accurate diagnosis whether viewed using the xVision browser on a workstation or on the mobile phone display (Table 2; Figs. 1 and 2). The only stated disadvantage of mobile phone consultations was the small size of the mobile phone display, which necessitated more zooming out, shifting, and rotating during the evaluation of some images. Evaluation of the location and type (stable/unstable) of injury was in full agreement between the workstation and mobile phone consultations. After consultation via mobile phone and workstation, 11 patients with unstable spinal injuries with or without cord lesions were transported to the Department of Neurosurgery at Bata Hospital for emergency treatment. Finally, the diagnoses were confirmed in all consulted patients by the Department of Neurosurgery staff.

Discussion

To the best of our knowledge, there is only one other study that compares the efficacy of consultation using a mobile phone with the efficacy of consultation at a standard hospital workstation.

To facilitate remote consultations, a network was developed in north Moravia (Czech Republic) between 2002 and 2003, which was initially based on Integrated Services Digital Network and later based on Asymmetric Digital Subscriber Line. The network enabled neurosurgeons to view medical images on home PCs by sharing the workstation screen images with the physician’s home PC via the IP addresses. It saved time and resources by reducing physicians’ hospital hours without undermining the quality

Fig. 1 Unstable injury of the cervical spine with a spinal cord lesion-comparison between workstation monitor and mobile phone display images.

of remote consultations. This system has since been extended to other medical departments that rely heavily on image transfer, such as the Department of Radiology.

Despite all of these advantages, the specialist would still have to remain close to a home PC that had a fixed Internet connection to the workstation at the hospital. At the turn of the

Fig. 2 Unstable injury of the lumbar spine without a spinal cord lesion-comparison between the workstation monitor and the mobile phone display images.

21st century, several articles appeared that recommended the increased use of mobile phone networks for medical image transfer [17].

Because of the increasing number of acute neurosurgery consultations in the Zlin region (approximately 100 spinal injuries per year) and after considering the technical and financial challenges [17], mobile consultation was added to the regional network. In 2006, we first tried to transfer and display images using the remote screen of a pocket digital assistant; but poor image quality and problems with data transfer without a fixed line discouraged use.

Improved mobile technology suggested that mobile phones with high-quality screens could be used to transmit images with sufficient resolution. During the testing phase from September 2009 to September 2010, we assessed the advantages and disadvantages of this method of consultation for the treatment of spinal injury patients admitted to regional hospitals. Compared with consultations through fixed- connection lines, the quality of the consultation was not markedly affected. The images were equally clear on the mobile phone display and on the workstation monitor.

We compared our study with the experiences of a Japanese group [17] that used a similar mobile phone-based remote consultation system but sent images through the multimedia messaging service. Using the multimedia messaging service system, there was a risk of incorrect image selection by a less experienced colleague or by the staff responsible for image transfer. Our method eliminated this risk. Although the images were uploaded by hospital staff, these individuals were not responsible for specific image selection. The selection and evaluation of the images were performed entirely by the consulting physician after all images from the patient were downloaded to the mobile phone. This scenario also excluded the possibility of sending incorrect images.

The most important advantage of this method was that it maintained the same quality of remote consultation as that provided by the fixed line system, but it enabled the physician to be mobile during half-service. The safety and security of the data were maintained by the method of transfer (VPN), the processing system (PACS), and the requirement of a unique user name and password.

One reported disadvantage was the small display size of the mobile phone, which necessitated zooming out and rotating the images during evaluation. It was also less comfortable for the user compared with the larger images viewed on the workstation. However, the main limitation of mobile phone consultation was the speed of data transfer (maximum rate, 512 kb/s). In the future, increased data transmission speeds and larger mobile phone displays with even higher resolution could further improve the quality of images transferred by mobile phone.

Conclusion

Remote medical consultation by transferring images and other relevant data over mobile phone networks was

sufficiently rapid and reliable for the accurate and Timely diagnosis of patients with spinal injuries. Indeed, the transferred CT or MRI images were of sufficient quality to allow diagnosis and decisions regarding further therapy. The time taken for this process was longer than the time required when consultations were performed using a hospital workstation running the xVision browser, but this method still allowed specialists to share their expertise with local or regional hospital staff.

Acknowledgments

The authors would like to thank the employees of Aura Servo (Praha, Czech Republic) for their logistical support during the implementation of this teleconsultation service.

References

1. Bailes JE. NeuroLink: a neurosurgical wide-area computer network. Neurosurgery 1994;35:732-6.
2. Batnitzky S, Rosenthal SJ, Siegel EL, et al. Teleradiology: an assessment. Radiology 1990;177:11-7.
3. Baxter KG, Wetzel LH, Murphey MD, et al. Wide area networks for teleradiology. J Digit Imaging 1991;4:51-9.
4. Gray WP. Report of a national neurosurgical teleradiology system based on optical scanning technology. Proceedings of the 9th European Congress of Radiology. (Abstract)Eur Radiol 1995;5:296.
5. Gray WP, Somers J, Buckley TF. Report of a national neurosurgical teleradiology system. J Telemed Telecare 1997;3(Suppl 1):36-7.
6. Gray WP, Somers J, Buckley TF. Report of a national neurosurgical emergency teleconsulting system. Neurosurgery 1998;42:103-8.
7. Filip M. Regional electronic consultation system. A Letter to the editor. Cesk Slov Neurol N 1998;5:62-3.
8. Spencer JA, Dobson D, Hoare M, et al. The use of a computerized image transfer system linking a regional neuroradiology center to its district hospitals. Clin Radiol 1991;44:342-4.
9. Jelasco DV, Southworth G, Purcell LH. Telephone transmission of radiographic images. Radiology 1978;127:147-9.
10. Lear JL, Manco-Johnson M, Feyerabend A, et al. Ultra-high-speed teleradiology with ISDN technology. Radiology 1989;171:862-3.
11. Filip M, Pasecny J, Valek J, et al. PC based teleconsulting in neurotrauma patients. Abstract) Book of abstracts of 12th World Congress of Neurosurgery, Sydney, Australia; 2001. p. 373.
12. Glowacki M, Czernicki Z, Jurkiewicz J, et al. Teleradiology in

neurosurgery, based on the experience of the Department of Neurosurgery, Polish Academy of Sciences. Pol J Radiol 2005;70: 27-34.

1. Templeton AW, Dwyer SJ, Rosenthal SJ, et al. A dial-up digital teleradiology system: technical consideration and clinical experience. AJR 1991;157:1331-6.
2. Walters KA. Telephoned head injury referrals: the need to improve the quality of information provided. Arch Emerg Med 1993;10:29-34.
3. Gentleman D, Jennett B, MacMillan R. Death in hospital after head injury without transfer to a neurosurgical unit: who, when, and why? Injury 1992;23:471-4.
4. Curtis DJ, Gayler BW, Gitlin JN, et al. Teleradiology: result of a field trial. Radiology 1983;149:415-8.
5. Yamada M, Watarai H, Andou T, et al. Emergency image transfer system through a mobile telephone in Japan: technical note. Neurosurgery 2003;52:986-90.