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

This Article Abstract Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Finkeissen, E. Articles by Wetter, T. Search for Related Content PubMed PubMed Citation Articles by Finkeissen, E. Articles by Wetter, T.

AIDA: Web Agents in Dental Treatment Planning

¹ Institute for Medical Biometry and Informatics, Department of Medical Informatics, University of Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, Germany; and
² Dental School, University of Heidelberg, Germany;

Correspondence: ^* corresponding author, Ekkehard_Finkeissen{at}med.uni-heidelberg.de

	Abstract

TOP Abstract Introduction The AIDA Methodology The Current AIDA Agents Discussion References

 
The objective of the AIDA project (Artificial Intelligent Dental Agents, http://aida.uni-hd.de) is the analysis of dental decision-making, the design of a computer-based decision support system, as well as the testing of the decision structure in interactions with dental experts, practicing dentists, and patients. The planning of the solution alternatives for an individual patient is based on a top-down structure for dental decision-making, aiming at a standardization of the argumentation. From a theoretical point of view, decision support can be provided only for anticipated decisions (planning). Moreover, only parts of these anticipated decisions can be supported. Accordingly, a separation of these partial aspects has to take place before one is able to build decision support systems. For prosthetic dentistry, clinicians have been shown how to use individual patient findings to sketch the possible treatment alternatives and later derive guidelines for the treatment. The planning module for fixed prostheses has already been integrated into a software agent. Planning modules for other types of prostheses are currently specified, implemented, and verified.

KEY WORDS: Dentistry • medical informatics • software • knowledge • ontologies • Internet • guideline

	Introduction

TOP Abstract Introduction The AIDA Methodology The Current AIDA Agents Discussion References

 
Dental treatment planning strongly depends on the educational background and the personal experiences of the dentist, as well the thoroughness and precision of the treatment planning. Therefore, the AIDA team developed a system for decision-making in dentistry, as part of the University of Heidelberg project to clarify these planning requirements (Finkeissen, 2002).

Prior approaches for dental treatment planning were restricted to partial aspects of a solution (Cordato, 1996; Han et al., 1998; Sellen et al., 1998; Zhang et al., 1998; Balch et al., 1999; Okumura et al., 1999; Pan et al., 1999). In contrast, AIDA’s methodology aims at a top-down partitioning of dental decision-making. Here, the separation of medical, technical, economic, and legal aspects organizes the multi-dimensional complexity of dentistry into small, easy-to-handle planning modules.

Before building a decision support system, one must clarify the parts of the dental tasks that have to be anticipated, i.e., planned. Moreover, not all parts of the planning can be automated. And computer support is useful only in cases of optimization of the overall treatment process. In other words, computer-based decision support can cover only specific parts of the dental tasks (Fig. 1).

View larger version (17K): [in this window] [in a new window]   Fig. 1 — Nesting in dental decision-making. Only parts of the amount of all dental tasks have to be planned (planning = anticipation of decisions). Moreover, only specific aspects of the planning can be supported by computer systems. Only optimized support is useful in the routine and should therefore be used in dentistry.

	The AIDA Methodology

TOP Abstract Introduction The AIDA Methodology The Current AIDA Agents Discussion References

 
AIDA starts planning the general goals for dental treatment and derives patient-specific target criteria from the patient’s individual findings. Based on this information, AIDA is supposed to plan all treatment alternatives meeting these target criteria. These treatment alternatives are presented as equivalents, which allow the practicing dentist to select the individually appropriate type of therapy. Here, "no prosthesis necessary" also represents a valid option in the decision-making process.

However, a decision methodology is abstracted from individual cases and cannot include all aspects of any individual patient. Accordingly, it is up to the dentist to select one of AIDA’s treatment suggestions and examine it in detail, if desired.

Since many of the fixed prosthetic solutions in dentistry involve more than one tooth, an appraisal of their necessity should be performed as early as possible; only this allows for the assessment of their effects on other teeth. After the possible prosthetic treatment alternatives have been determined, detailed questions can be clarified to test the concrete applicability of the desired prosthetic solution (i.e., periodontal requirements).

A comparison of AIDA treatment planning with real-life patient records has already yielded valuable indications for extensions and adaptations related to the planning rules (Cordato, 1996). The current version of AIDA includes specific planning aspects of restorative dentistry which are related to the planning of a prosthetic construction. From this extension, a further increase in both recall and precision is expected in AIDA’s planning.

	The Current AIDA Agents

TOP Abstract Introduction The AIDA Methodology The Current AIDA Agents Discussion References

 
The AIDA interface supports the anonymous coding of the findings. The information is sent to the AIDA agent, where the actual planning is carried out. In the planning system, the transferred findings are checked for integrity according to plausibility rules. Only after the integrity of the transferred information has been established can AIDA suggest plans.

The transfer of findings is currently being adapted to a new standard established within the AIDA project. This new standard is aiming at a concise coding of the patient status, with a current emphasis on information about the set of teeth.

After re-implementation, the AIDA planning suggestions can be accessed both via Web browser and by other machines (e.g., dental billing software) (Fig. 2).

View larger version (24K): [in this window] [in a new window]   Fig. 2 — Revised documentation abbreviations will make dentists’ communication more precise. Implemented in XML, the information can be both automatically processed and appropriately presented for the respective user group.

• How many solutions are there?
  
• How can they be described?
  
• What do they look like?
  

In the next versions, the AIDA agents will be able to provide information on further questions:

• What is the rationale for the decision-making of each solution?
  
• How can a guideline for the implementation of the treatment be provided?
  
• How much does the solution cost?
  

In the latest version, AIDA’s output is provided in XML format and can be visualized in a human-readable form by style sheets. Thus, both humans and machines can use AIDA’s planning, where the styles can be adapted to the user roles (expert, practitioner, patient).

The solutions provided by AIDA can, therefore, be passed along to a billing program and shown on the planning form. The dentist can accept one of the suggestions and edit it as desired. The treatment planning for fixed prostheses can soon be planned online and for free (Fig. 3).

View larger version (75K): [in this window] [in a new window]   Fig. 3 — The Web interface of the AIDA agent for dentists. For subsequent checking of appropriateness of the treatment alternative, the findings and their interpretation are also included in the planning output.

	Discussion

TOP Abstract Introduction The AIDA Methodology The Current AIDA Agents Discussion References

 
The computer cannot replace the dentist. A machine cannot make decisions on its own, because it does not understand the patient’s problems and needs in every detail. For this reason, computer-assisted decision-making in the future will remain a communication medium between experts and dentists. As a neutral authority, however, such a communication instrument can promote rational discussions among experts as well as between dentists and patients. Thus, it can help explain the necessity of, e.g., complex treatment alternatives.

During the development of AIDA, it became clear that a uniform and structured method of documentation for the comprehensible communication of dental decisions is essential. Fundamental deficits in the practice of dental documentation were revealed (Finkeissen et al., 2002a). Currently, the AIDA team includes further specifications for dental documentation.

A nationwide expansion of the dental expert team will hopefully critique AIDA’s output, thereby providing support for planning precision as well as the integration of a broader range of treatment alternatives into the AIDA methodology. As soon as AIDA includes a convincing number of treatment alternatives, its structure can also serve the definition of comparative studies on the evaluation of the individual treatment alternatives that can later be included into the AIDA approach.

	Acknowledgments

 
The development of this manuscript was supported in part by the Medical Faculty at the University of Heidelberg (Project 25/1996).

	Footnotes

 
Publication supported by Software of Excellence (Auckland, NZ)

	References

TOP Abstract Introduction The AIDA Methodology The Current AIDA Agents Discussion References

 
Balch DC, Warner DJ, Gustke SS (1999). Highlights from the Third Annual DMI Conference—medical knowledge on demand. MD-Computing (March/April):48–50.

Cordato M (1996). A mathematical study of anterior dental relations: Part II, Incisor and canine overjet. Aust Orthod J 14(3):143–149.[Medline]

Eisner J (1999). The future of dental informatics. Eur J Dent Educ 3:61–69.

Finkeissen E (2002). Zahnmedizinische Entscheidungsstrukturen—Grundlagen einer wissensbasierten Zweitmeinung. In: Institut für Medizinische Biometrie und Informatik, Abteilung Medizinische Informatik. Heidelberg: Universität Heidelberg.

Finkeissen E, Hassfeld S, Koke U, Helmstetter-Blasius C, Eickholz P, Wetter T (2002a). Problems with current dental documentation in Germany. Heidelberg, Germany: Institute for Medical Biometry and Informatics, Department of Medical Informatics, p. 15.

Finkeissen E, Koke U, Hassfeld S, Helmstetter C, Weber R, Wetter T (2002b). AIDA—iterative approach for the improvement of a knowledge-based decision methodology in dentistry. Quintessence Int (submitted).

Han K, Lu R, Ma Y, Lu P, Zhang H, Wang W (1998). Reconstruction of three-dimensional digital teeth. Chin J Dent Res 1(2):22–25.[Medline]

Haux R (1999). Medizinische Informatik. In: Medizinische Physik. Vol. 1. Bille J, Schlegel W, editors. Heidelberg: Springer, pp. 459–484.

Okumura H, Chen L, Yokoe Y, Tsutsumi S, Oka M (1999). CAD/CAM fabrication of occlusal splints for orthognathic surgery. J Clin Orthod 33:231–235.[Medline]

Pan S, Yin Y, Feng H (1999). Three-dimensional finite element analysis and comparison of stress distribution in overdentures supported with bar attachments and telescopic crowns. Chin J Dent Res 2(1):21–30.

Sellen P, Jagger D, Harrison A (1998). Computer-generated study of the correlation between tooth, face, arch forms, and palatal contour. J Prosthet Dent 80:163–168.[Medline]

Zhang X, Yang C, Gao J, Ding Z, (1998). Computer-assisted design and manufacture of posterior full crowns. Chin J Dent Res 1(3):30–34.[Medline]

This Article Abstract Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Finkeissen, E. Articles by Wetter, T. Search for Related Content PubMed PubMed Citation Articles by Finkeissen, E. Articles by Wetter, T.