Simulated surgery on computed tomography and magnetic resonance images: an aid for intraoperative radiotherapy

Intraoperative radiotherapy (IORT) is a relatively new technique in which irradiation with electrons is performed during an open surgery procedure. This approach poses significant problems in obtaining accurate dosimetry, since neither the pre- nor the postoperative patient images actually matches the irradiation field. Our objective was to implement a software tool able to provide an estimate of the dose distribution, overcoming the problem of the geometrical mismatch between the images and the surgical field during the irradiation. The program was developed in the C programming language, on a noncommercial version of a Philips EasyVision workstation. The application allows to create a new data set by manipulating the preoperative computed tomography and magnetic resonance images in order to simulate the final geometry of the surgical area during the IORT procedure. The exact dose distribution can then be calculated by transferring these new images to a standard radiotherapy planning system. Also an approximate dose distribution can be quickly displayed by superimposing isodose curves obtained from a water phantom. The proposed approach introduces a helpful tool for dosimetry and planning in IORT protocols, improving their accuracy and safety and allowing for more objective quality control and patient follow-up.