Reply to Heatley and Bell

The letter to the Editor by Dr. Heatley and Prof. Bell refers to an article published by our group in Endoscopy several months ago [1]. On the basis of studies that were published between 1993 and 1998, they argue that it is inappropriate in feasibility studies of tele-endoscopy to focus exclusively on the picture quality. They state that it is ”important to take in wider considerations, such as cost, flexibility, availability, medical practices, audio requirements, etc.”. In our view, however, a new technology should initially be evaluated in high-quality feasibility studies that examine the focal point, and secondly in comparative studies to demonstrate its equality or superiority to existing techniques. In addition, it is important to take into account appropriate clinical applications and the demand for a new technique.

The major issue in video endoscopy is picture quality. In recent years, all endoscope manufacturers have significantly increased the resolution of the video endoscopes that are available. In the near future, it will become possible to visualize the blood flow in mucosal capillaries, using endoscopes with 1 million pixels and 200-fold magnification. High-resolution video endoscopy and chromoendoscopy are already allowing today’s endoscopists to identify flat adenomas, intraepithelial neoplasia, and early carcinoma which were too often overlooked in the past due to inadequate image resolution and also probably because of low attention to such conditions or inexperience on the part of endoscopists [2]. Since these lesions are curable by local endoscopic therapy [3], requirements in video endoscopy examinations have rapidly changed in recent years. We therefore consider that it is not only appropriate but also necessary to focus on picture quality when analyzing tele-endoscopy.

We therefore carried out a prospective study using a highly sophisticated and technically well-defined system for digital video transmission [4]. The participating endoscopists were asked to give a rating for whether they would be able to identify all mucosal lesions on basis of the picture quality displayed which was also quantified in the study. We concluded from the results that endoscopic video material transmitted at 40 megabit/s in Moving Pictures Experts Group 2 (MPEG2) [4 : 2:2] format has unlimited diagnostic validity and can therefore be used for real-time telemedical applications without restriction. The technical equipment was then used to modify a variety of parameters in a randomized fashion within a wide range: compression algorithms, network bandwidths, and impairments due to network overload situations. The evaluation of the video material by experienced endoscopists turned out to be surprisingly uniform, and the test subjects were evidently capable of recognizing and rating technically objective changes in the quality of the video picture. Finally, all tested technical parameters had a significant impact on the quality of the video picture, and the resulting picture quality subsequently influenced the usability of the images for diagnostic purposes. In summary, real-time digital video transmissions using MPEG2 [4 : 2:2] at 40 megabit/s and the asynchronous transfer mode (ATM) protocol were not distinguishable from the original picture in our clinical setting, but any experimental variation of the transmission parameters (bandwidth, method of compression, impairments) led to a noticeable reduction in picture quality and to a decrease in diagnostic usability. Although these findings are rather in contrast to the results of most telemedical studies, which have started from the poorest picture quality and risen not much higher, we demonstrated by using a ”quality-standard” (40 megabit/s in MPEG2 [4 : 2:2] format) and a ”step-down” methodology that the threshold for detectable diagnoses and unacceptable picture quality in video endoscopy lies much higher than previously described.

As we pointed out, the picture quality is the primary technical prerequisite in video endoscopy, and of course real-time tele-endoscopy also has to meet this requirement. All other issues in the discussion are therefore secondary, although not unimportant.

Heatley and Bell state that training and supervision are important applications for telemedicine, allowing a specialist to be made available at a remote site; so far, so good. ”Stationary” or ”frozen” images or ”full resolution within a second or so” produced an image quality ”very close to the original” in their experience. Otherwise, they state that improved voice quality will be needed for future applications. Evidently, however, a technically defined latency for real-time video transmission was not on their list of requirements. Since the time loss during transmission (latency) doubles in bidirectional communications, doctors at the Department of Surgery at the University of Erlangen-Nuremberg [5] (http://www.meditos.de/chirurgie/telemedizin/), who carried out a quality control study of cancer surgery using the system we tested (200 ms for bidirectional communication, including coding, compression, transmission, decoding), offered another example of requirements regarding latency: ”You are standing at the remote site and I begin to open your mother’s abdomen with a scalpel; I will stop cutting when I hear you saying ‘stop.’ ” Although this example is slightly exaggerated, it demonstrates another important issue in telemedicine - that technical requirements need to be adapted for each individual application. Many possible applications might require a high video quality, but not real-time transmission. Not all applications need full color depth or full image resolution, while others do. As we stated, an asynchronous teleconsultation appears to be a good option if the receiver’s decision is not needed at once (i. e., for data transfer to a specialist or for quality control) [1]. This would facilitate the accessibility of specialists and decrease expenses for technical requirements. Considerations, such as costs, flexibility, availability, etc. are of course important, but they have their place in concrete situations in which a specific application is to be used. Our exploratory approach on real-time tele-endoscopy, however, was more general. We demonstrated the theoretical difficulties with image quality arising from coding, transmission, and decoding of digital video streams. The results focus on real-time transmissions and do not justify recommendations for or rejections of other applications than those using real-time transmission.

Finally, there is the issue of whether telemedicine is a ”solution looking for problems”. Where is the need for it, and what are the clinical applications? Wootton described two situations justifying the use of telemedicine in the future [6]: firstly, when there is no alternative, and secondly, when it is in some way better than traditional medicine. There are some situations in which it is difficult or impossible to transport a doctor to a patient, or vice versa (navigation, aviation, astronautics, and war); however, these situations are rare and are not of general interest. So what are the theoretical advantages of telemedicine? Telemedicine could facilitate and improve the usual process of communication between doctors and patients, physicians and specialists. In this situation, ”telemedicine” may reflect a continuous process of substitution of older communication techniques by more recent ones (letter, fax, e-mail, remote electronic access). However, ”medicine at a distance” [7] will include extension of the doctor’s eyes, ears, and other sense organs - even his hands - for remote procedures, such as teleradiology, telepathology, teleconsulting, tele-endoscopy, and telesurgery. Telemedicine in this sense is a substitution for traditional medical techniques. Current users of telemedicine therefore need to confirm the efficiency and efficacy of ”experimental” applications in comparison with traditional medical techniques. Reasons for looking for new techniques may include assumed deficiencies of currently used processes, and the hope that telemedicine may be able to improve the situation: the need to move the patient or the doctor, a local absence of certain specialties and skills, as well as hygienic and economic considerations. Given these aims, basic evaluations of the feasibility of telemedical applications, controlled trials that allow generalization of results, and finally rigorous calculation of the costs and benefits are warranted [8] [9]. At present, however, there is little evidence that telemedicine is clinically effective [6].

References

• 1 Rabenstein T, Maiss J, Naegele-Jackson S. et al . Tele-endoscopy: influence of data compression, bandwidth and simulated impairments on the usability of real-time digital video endoscopy transmissions for medical diagnoses.  Endoscopy. 2002;  34 703-710
• 2 Saitoh Y, Waxman I, West A B. et al . Prevalence and distinctive biologic features of flat colorectal adenomas in a North American population.  Gastroenterology. 2001;  120 1657-1665
• 3 May A, Gossner L, Pech O. et al . Local endoscopic therapy for intraepithelial high-grade neoplasia and early adenocarcinoma in Barrett’s oesophagus: acute-phase and intermediate results of a new treatment approach.  Eur J Gastroenterol Hepatol. 2002;  14 1085-1091
• 4 Naegele-Jackson S, Holleczek P, Rabenstein T. et al .Influence of compression and network impairments on the picture quality of video transmissions in tele-medicine. In: Sprague RH Jr, (ed) Proceedings of the 35th Hawaii International Conference on System Sciences (HICSS-35’02), 7-10 January 2002,. Big Island, Hawaii. Los Alamitos; CA: Institute of Electrical and Electronics Engineers (IEEE) Computer Society, 2002 : 159b (http://www.computer.org/proceedings/hicss/1435/volume6/14350159babs.htm) 2002
• 5 Weber H, Schick C, Horbach T. et al . Zukunft der Telemedizin - aus der Sicht des Chirurgen [in German].  Endosk Heute. 2001;  1 54
• 6 Wootton R. Telemedicine: a cautious welcome.  BMJ. 1996;  313 1375-1377
• 7 [Editorial] . Telemedicine: fad or future?.  Lancet. 1995;  345 73-74
• 8 Wyatt J C. Telemedicine trials: clinical pull or technology push?.  BMJ. 1996;  313 1380-1381
• 9 Murdoch I. Telemedicine.  Br J Ophthalmol. 1999;  83 1254-1256