An intelligent oncology workstation for the 21st century

The 21st century will see the routine clinical use of improved radiotherapy treatment techniques, such as intensity modulated radiation therapy (IMRT). An essential feature for the implementation of IMRT is the ability to identify precisely the location of body structures, particularly those involved with cancer and surrounding uninvolved regions, which should receive minimal dose. The aim of this paper is to introduce the concept of an intelligent oncology workstation: a one-stop workstation combining the automatic delineation of structures of interest with the optimization and scheduling of radiotherapy treatment delivery. M a t e r i a l a n d m e t h o d s. Software tools have been programmed using the Matlab programming environment and C/C++. Rule based algorithms refine contours obtained using low-level image processing tools. Treatment planning optimization algorithms combine a least square methodology with multiple objective genetic algorithms. R e s u l t s. The imaging software can outline successfully regions of interest such as the bladder, rectum and pelvic bones. Optimization provides a set of solutions for coplanar beam orientation and modulation in intensity, taking into account dose delivery constraints. C o n c l u s i o n s. Using clinical experience as well as the raw image data, a physician employing the workstation can improve IMRT planning with automatic identification of body organs and structures. Multiple objective genetic algorithms (MOGA) exploiting the concept of Pareto optimality offer advantages over the traditional weighted sum approach. MOGA provides clinicians with a set of equally good IMRT plans that take into account practical limitations of the treatment delivery mechanisms.