gms | German Medical Science

GMS Current Topics in Computer and Robot Assisted Surgery

Deutsche Gesellschaft für Computer- und Roboterassistierte Chirurgie (CURAC)

ISSN 1863-3153

UML based modeling of medical applications workflow in maxillofacial surgery

Research Article

  • corresponding author Monica Toma - Department of Oral and Cranio-Maxillofacial Surgery, University of Heidelberg, Heidelberg, Germany
  • Alexander Busam - Department of Oral and Cranio-Maxillofacial Surgery, University of Heidelberg, Heidelberg, Germany
  • Tobias Ortmaier - Institute of Robotics and Mechatronics, DLR - German Aerospace Center, Wessling, Germany
  • Jörg Raczkowsky - Institute for Process Control and Robotics (IPR), University of Karlsruhe, Karlsruhe, Germany
  • Claudia Höpner - Department of Oral and Cranio-Maxillofacial Surgery, University of Heidelberg, Heidelberg, Germany
  • Rüdiger Marmulla - Department of Oral and Cranio-Maxillofacial Surgery, University of Heidelberg, Heidelberg, Germany

GMS CURAC 2007;2(1):Doc03

Die elektronische Version dieses Artikels ist vollständig und ist verfügbar unter:

Veröffentlicht: 14. März 2007

© 2007 Toma et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen ( Er darf vervielfältigt, verbreitet und öffentlich zugänglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.


This paper presents our research in medical workflow modeling for computer- and robot-based surgical intervention in maxillofacial surgery. Our goal is to provide a method for clinical workflow modeling including workflow definition for pre- and intra-operative steps, analysis of new methods for combining conventional surgical procedures with robot- and computer-assisted procedures and facilitate an easy implementation of hard- and software systems.


Medical applications workflow refers to the operational aspect of clinical processes, partly or wholly automated, detailing the structure of tasks, their relative order, synchronization, actors, and information flows, according to a set of procedural rules. Modeling medical applications workflow in maxillofacial surgery is of paramount importance for ensuring an accurate outcome, both aesthetically and functionally. Moreover, it has been shown that workflow models facilitate the transfer of know-how about the new system among the participants and also stimulate discussion about the new work processes. For this, however, workflow modeling needs a language that is intuitive and easy to use by all the participants. Our objective is to show that clinical workflow can be modeled using a mainstream business modeling language such as Unified Modeling Language (UML) 2.0 [1].

UML activity diagrams

The Unified Modeling Language (UML) is a family of graphical notations, backed by a single meta-model, that helps to describe and design software systems, particularly software systems built using the object-oriented (OO) style. UML was developed by the Object Management Group (OMG) as an open standard and was initially conceived as a general-purpose language for modeling object-oriented software applications. We use UML 2.0, which compared to UML 1.x has introduced some changes in the semantics. Activity diagrams now correspond widely to the Petri semantics instead of state machines.

For system analysis UML offers different diagram-types such as use case diagrams or activity diagrams. The use case view models the functionality of the system as perceived by outside users, called actors. A use case is a coherent unit of functionality expressed as a transaction among actors and the system. The purpose of the use case view is to list the actors and use cases and show which actors participate in each use case. UML activity diagrams, which represent operational workflows of a system, are widely used as a representation language for workflows, particularly for software development and business process modeling [2]. They describe the dynamic behavior of a system. In Activity Diagrams, the actions to be realized are identified, as well as the relationships among them. It is also possible to identify the objects involved in each activity and to define changes in their role, state and attribute values.

The main concepts of Activity Diagrams are shown in Table 1 [Tab. 1].

In a first step of system specification, the system is described in a main scenario as a sequence of numbered steps in order to understand its functional requirements [3]. It is then modeled with a UML use case diagram. In the next step we model the behavior of the system in an activity diagram, which is similar to a flowchart, but the difference is that it supports parallel behavior. This allows the optimization of the workflow by parallelizing those activities that are accomplished sequentially by different actors (as in Figure 3).

Case studies

Hand-guided laser scanner registration

The first case we investigated is the modeling of the workflow for the hand-guided laser scanner registration. The scanner is used for image-to-patient registration in the operation theatre [4]. It provides the initial position and orientation of the patient for navigated maxillofacial surgery [5]. The accuracy of the registration has much influence on the total accuracy of the navigation system. The head of the patient is scanned intraoperatively and the acquired points are matched with the surface data of the CT/MRT acquired preoperatively. To compensate movements of the patient’s head during the scan-process a tracking body is attached to the patient’s maxilla.

The clinical workflow of the hand-guided laser scanner registration is modeled in users’ view in the activity diagram in Figure 1 [Fig. 1]. Actors (Dentist, Maxillofacial surgeon, etc.), that are identified in the use cases, are grouped in different responsible areas. Further the timeline of the whole process is separated in pre-, intra- and postoperative steps.

To keep the main activity diagram of the clinical workflow simple, complex activities are decomposed in a sub-activity diagram as in Figure 2 [Fig. 2].

Robot-assisted dysgnathia surgery

The second case is concerned with the robot assisted dysgnathia surgery workflow (Figure 3 [Fig. 3]). Dysgnathia is a congenital anomaly, namely, an anomaly in the form and position of the jaw resulting from growth disturbances in the facial skull. In order to correct this problem a translocation of bone segments is necessary. However, the precision achieved in the planning phase is usually not translated to patients. The aim of robot assisted surgery is to combine the surgeon skill and experience with the high precision of a robot in order to carry out the bone repositioning.

During a robot-assisted dysgnathia operation, the surgeon is supported by a complex system which uses a preoperative treatment plan [6], an infrared navigation system for monitoring both the patient and the robot, and a surgical robotic system. The robot is used as a tool for registration and for maintaining the desired bone position. In the intra-operative setting, after the registration is completed and until the bone is repositioned, the surgeon performs the surgery in conventional mode.

Surgical planning being a complex activity is decomposed in a sub-activity diagram (Figure 4 [Fig. 4]).


Clinical workflow analysis and modeling play an important role in computer- and robot-based surgical intervention in maxillofacial surgery. For the implementation of a new system, an understanding of the processes involved is necessary. Workflow models can help to determine the impact of a new system on the clinical workflow; define the expectations and requirements for a new system; and manage the change process associated with the implementation of a new system.

There are several benefits related to the use of UML for the description and specification of medical workflow. First of all, it is expressive, being clear and easy to understand for all participants, from the system engineers to the surgeon. It is also suitable for translation into mathematical and computational formalisms, which enables analysis. Moreover, it is a freely available official standard and benefits from the availability of advanced modeling tools.



This work was supported by the Deutsche Forschungsgemeinschaft within the “Oberflächenbasierte Patientenregistrierung mit einem handgeführten Laserscanner” project and the “Räumlich präzise Repositionierung von Knochen in sechs Freiheitsgraden bei der Gesichtsschädelrekonstruktion” project.

Conflicts of interest

None declared.


OMG. Unified Modeling Language Specification v 2.0. Object Management Group; 2004.
Rumbaugh J, Jacobson I, Booch G. The Unified Modeling Language Reference Manual. 2nd ed. Addison Wesley; 2004.
Bastos RM, Dubugras D, Ruiz A. Extending UML Activity Diagram for Workflow Modeling in Production Systems. Proceedings of the 35th Hawaii International Conference on System Sciences, 2002.
Suppa M, Hirzinger. A Novel System Approach to Multisensory Data Acquisition. 8th Int. Conference on Intelligent Autonomous Systems, IAS-8, Amsterdam, 2004.
Marmulla R, Hassfeld S, Lueth R, Muehling J. Laser-scan-based navigation in cranio-maxillofacial surgery. J Craniomaxillofac Surg. 2003;31:267-77.
Schorr O, Raczkowsky J, Woern H. Supporting Operation Planning by Workflows. Proceedings of the 18th International Congress and Exhibition on Computer Assisted Radiology and Surgery; 2003.