HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Dennis, M. J. Articles by Dammann, F. Search for Related Content PubMed Articles by Dennis, M. J. Articles by Dammann, F. Related Collections Related Article

Invited Commentary • Author's Response

Department of Radiology, Medical College of Ohio, Toledo

	Commentary

Top Commentary Response

 
The preceding article by Dammann et al discusses and illustrates the use of digital systems in planning, simulating, and even actively facilitating complex surgical procedures. It highlights some of the potential benefits as well as some of the challenges of putting computer-aided surgical planning into practice.

Graphical VR displays are shown that allow the test fitting of an implantable hearing aid from CT data of a cadaveral specimen, with the computer providing collision or interference detection. As Dammann et al noted, a major benefit of using the planning software is the ability to identify cases in which the surgical procedure is not likely to be successful or in which an alternative approach is needed due to the patient's anatomy. It is beyond the scope of the authors' feasibility study to determine whether the clinical benefit of this type of analysis warrants its use for this application.

The quality of performance on the patient data in this study was disappointing but should improve with the development of alternative surface generation algorithms, interpolation methods, and segmentation tools. For example, many of the commercial CT systems make use of surface generation algorithms, which estimate a boundary position between pixels for improved results (Fig 3 in the preceding article ) rather than translating the data to the binary structure (Fig 5 in the article) prior to surface generation. The quality of the analysis is only as good as that of the input data, which should help determine the scanning parameters. As seen in the study by Dammann et al, this can mean acquiring closely spaced, high-resolution images and processing large quantities of information. Caution in the interpretation and use of this reconstructed data is also in order due to potential problems in the 3D surface renderings of bone structures. Partial volume averaging may alter structural dimensions or cause thin or trabecular bone structures to disappear.

Several different software packages were used in the study to obtain the graphical results. This not only causes differences in the user interfaces and makes use more difficult; it also necessitates data translation and communication between packages and workstations. The authors' basic "test fit" approach may be useful but does not allow the surgeon to modify the anatomy to make the device fit. Future integrated implementations will likely include additional surgical simulation capabilities.

The incorporation of useable surgical planning and simulation tools will be very important. Equally important, however, are the personnel involved in this procedure. One needs to understand the technical system for the transfer of information and the operation of the interactive software packages, be familiar with the imaged anatomy and the limitations of the images, and have an understanding of the surgical procedure and its requirements. The manual segmentation procedure used in the study by Dammann et al required approximately 45 minutes of work by an experienced radiologist. Additional time is required for simulation of the implantation procedure. Even with the anticipated improved efficiency that should accompany technical advances in this area, a number of questions are apparent regarding cost and benefit, efficient use of time and resources within the radiology department, and reimbursement mechanisms. It will be important to identify current surgical problems that can effectively be addressed with this technical solution, making these efforts worthwhile.

Local logistical issues include determining who acquires the necessary hardware and software and defining the respective roles of the radiologist and surgeon in acquiring and processing the data. If the radiologist is to be a primary consultant in the use of image data for surgical planning rather than just the supplier of raw images, he or she will need additional technical and clinical skills (along with a strong interest in the process). Advanced capabilities such as those described by Dammann et al will continue to have an impact on the role of the radiologist and of associated imaging science resources in the clinical environment.

Author's Response

Department of Diagnostic Radiology, University Hospital Tübingen, Tübingen, Germany

	Response

Top Commentary Response

 
Digital surgical planning procedures based on radiologic data are increasingly being used in a variety of surgical subspecialties. Reimbursement for these systems (eg, stereolithographic modeling, intraoperative navigation tools, surgical robot systems), parts of which are very expensive, seems to be handled by the surgical units. Radiologists are usually not involved in the rapid developments in this promising field. However, a special radiologic competence is demanded regarding potential problems in the use of digital image postprocessing procedures, as Dr Dennis points out in his commentary . A radiologist who takes charge of the whole data acquisition and postprocessing procedure will be best equipped to guarantee the quality of constituent procedures in the clinical setting. The radiologist who denies his or her responsibility in this field due to lack of financial reimbursement or for some other reason will instantly be excluded from this part of the process and may also be at risk for claims involving the image-generating modalities, including CT or MR imaging, which represent the important first step of the process.

With the simulation of surgical procedures, the radiologist must be familiar with some new technical and clinical details (as with many other radiologic techniques that have been developed in recent years). Furthermore, greater cooperation with the surgical team is crucial.

The manual segmentation of CT data, which is required to simulate the excavation of the mastoid cells, is an essential task for the radiologist. With our application, there are no surgical variations to consider because the removal of the mastoid cells is always complete. This may be different with other applications.

The reimbursement dilemma may be mitigated by the use of software tools that are widely accessible, moderately priced, and easy to use and that run on standard PC-based hardware that may already be at hand. Unlike with other systems, these criteria have been major considerations in our attempt to improve acceptance of the procedure in the clinical environment.

Related Article

Computer-aided Surgical Planning for Implantation of Hearing Aids Based on CT Data in a VR Environment

Florian Dammann, Andreas Bode, Erwin Schwaderer, Michael Schaich, Martin Heuschmid, and Marcus M. Maassen
RadioGraphics 2001 21: 183-190. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Dennis, M. J. Articles by Dammann, F. Search for Related Content PubMed Articles by Dennis, M. J. Articles by Dammann, F. Related Collections Related Article