Iterative training of an automated dosimetric planning model for stereotactic radiotherapy in thoracic and lumbar spine pathologies

The study presents the results of applying an iterative approach to optimizing automated multi-criteria ionizing radiation dose distribution models for stereotactic spine radiotherapy. Two models were used: SRT_Vert_BM (Baseline Multicriteria Model) and SRT_Vert_IM (Iterative Multicriteria Model). The former was built on a clinical dataset of dosimetric plans, while the latter was created using an iterative method where the training dataset was formed from plans generated by the baseline model. Validation demonstrated that SRT_Vert_IM provides statistically significant improvements in predictive performance (increased coefficient of determination R², reduced mean squared error MSE) and dosimetric parameters for bilateral parenchymal organs at risk (p < 0.05) while maintaining clinically acceptable target volume coverage. Iterative training enhanced the reproducibility and reliability of dosimetric plan quality, confirming the potential of this approach for advancing automated treatment planning.

1. Teoh M., Clark C.H., Wood K., Whitaker S., Nisbet A. Volumetric modulated arc therapy: a review of current literature and clinical use in practice. The British Journal of Radiology, 2011, vol. 84, no. 1007, pp. 967–996. https://doi.org/10.1259/bjr/22373346

2. Schubert C., Waletzko O., Weiss C., Voelzke D., Toperim S., Roeser A., Puccini S. [et al.]. Intercenter validation of a knowledge based model for automated planning of volumetric modulated arc therapy for prostate cancer. The experience of the German RapidPlan Consortium. PLoS One, 2017, vol. 12, no. 5, art. ID e0178034. https://doi.org/10.1371/journal.pone.0178034

3. Wang J., Hu W., Yang Z., Chen X., Wu Z., Yu X., Guo X. [et al.]. Is it possible for knowledge-based planning to improve intensity modulated radiation therapy plan quality for planners with different planning experiences in left-sided breast cancer patients? Radiation Oncology, 2017, vol. 12, no. 1, art. ID 85. https://doi.org/10.1186/s13014-017-0822-z

4. Scaggion A., Fusella M., Roggio A., Bacco S., Pivato N., Rossato M. A., Peñaa L. M. A., Paiusco M. Reducing inter- and intra-planner variability in radiotherapy plan output with a commercial knowledge-based planning solution. Physica Medica, 2018, vol. 53, pp. 86–93. https://doi.org/10.1016/j.ejmp.2018.08.011

5. Lahanas M., Schreibmann E., Baltas D. Multiobjective inverse planning for intensity modulated radiotherapy with constraint-free gradient-based optimization algorithms. Physics in Medicine and Biology, 2003, vol. 48, no. 17, pp. 2843–2871. https://doi.org/10.1088/0031-9155/48/17/307

6. Pietkevich M. N., Yushkevich V. Yu. Development and machine learning of a multi-criteria ionizing radiation dose distribution model in the Eclipse treatment planning system. Vestsi Natsyyanal’nai akademii navuk Belarusi. Seryya fizikatekhnichnykh navuk = Proceedings of the National Academy of Sciences of Belarus. Physical-technical series, 2025, vol. 70, no. 2, pp. 166–176 (in Russian). https://doi.org/10.29235/1561-8358-2025-70-2-166-176

7. Shiraishi S., Moore K.L. Knowledge-based prediction of three-dimensional dose distributions for external beam radiotherapy. Medical Physics, 2016, vol. 43, no. 1, art. ID 378. https://doi.org/10.1118/1.4938583

8. Monz M., Küfer K. H., Bortfeld T. R., Thieke C. Pareto navigation – algorithmic foundation of interactive multicriteria IMRT planning. Physics in Medicine and Biology, 2008, vol. 53, no. 4, pp. 985–998. https://doi.org/10.1088/00319155/53/4/011

9. Fogliata A., Wang P. M., Belosi F., Clivio A., Nicolini G., Vanetti E., Cozzi L. Assessment of a model based optimization engine for volumetric modulated arc therapy for patients with advanced hepatocellular cancer. Radiation Oncology, 2014, vol. 9, art. ID 236. https://doi.org/10.1186/s13014-014-0236-0

10. Fogliata A., Reggiori G., Stravato A., Lobefalo F., Franzese C., Franceschini D., Tomatis S. [et al.]. RapidPlan head and neck model: the objectives and possible clinical benefit. Radiation Oncology, 2017, vol. 12, no. 1, art. ID 73. https://doi.org/10.1186/s13014-017-0805-0

11. Fogliata A., Nicolini G., Clivio A., Vanetti E., Laksar S., Tozzi A., Scorsetti A. C., Cozzi L. A broad scope knowledge based model for optimization of VMAT in esophageal cancer: validation and assessment of plan quality among different treatment centers. Radiation Oncology, 2015, vol. 10, art. ID 234. https://doi.org/10.1186/s13014-015-0545-y

12. Foy J. J., Marsh R., Ten Haken R. K., Younge K. C., Schipper M., Sun Y., Owen D., Matuszak M. M. An analysis of knowledge-based planning for stereotactic body radiation therapy of the spine. Practical Radiation Oncology, 2017, vol. 7, no. 5, pp. 355–360. https://doi.org/10.1016/j.prro.2017.02.007

13. Li N., Carmona R., Sirak I., Kasaova L., Followill D., Michalski J., Bosch W. R. [et al.]. Highly efficient training, refinement, and validation of a knowledge-based planning quality-control system for radiation therapy clinical trials. International Journal of Radiation Oncology – Biology – Physics, 2017, vol. 97, no 1, pp. 164–172. https://doi.org/10.1016/j.ijrobp.2016.09.022

14. DVH Estimation Algorithm for RapidPlan. Eclipse Photon and Electron Algorithms Reference Guide. Palo Alto, CA, Varian Medical Systems, 2014, pp. 219–229. Available from: https://www.varian.com/sites/default/files/resource_attachments/EclipsePhotonElectronAlgorithmsRef_0.pdf (accessed: 1 February 2025).

15. El Naqa I., Ruan D., Valdes G., Dekker A., McNutt T., Ge Ya., Wu Q. J. [et al.]. Machine learning and modeling: Data, validation, communication challenges. Medical Physics, 2018, vol. 45, no. 10, pp. 834–840. https://doi.org/10.1002/mp.12811