Enhancing precision in proton therapy: Utilizing machine learning for predicting Bragg curve peak location in cancer treatment

Abstract

In proton beam therapy, the Bragg peak is the point where protons lose energy the fastest. This point is crucial for dose control, preserving healthy tissues, minimizing lateral scattering, and the success of treatment planning. However, accurately predicting the location of the Bragg peak is challenging due to the complex interactions of protons with tissues. This study proposes a machine learning (ML) approach to predict the exact location of the Bragg peak from phantom tissue proton beam therapy experiments. A dataset comprising the eight most commonly used biomaterials, which mimic human tissue in proton therapy procedures, has been curated for this study. Various ML models are benchmarked to find the most successful approach. ML model parameters are further optimized using a metaheuristic approach to achieve the highest prediction capability. In addition, feature contributions of each feature in the dataset are analyzed using an explainable artificial intelligence (XAI) technique. According to experimental results, Random Forest (RF) model that is optimized with Genetic Algorithm (GA) achieved 0.742 Correlation Coefficient (CC) value, 0.069 Mean Absolute Error (MAE) and 0.145 Root Mean Square Error (RMSE) outperforming other ML models. The proposed approach can track and predict the movement of the proton beam in real-time during treatment, enhancing treatment safety and contributing to the more effective management of the treatment process. This study is the first to predict exact Bragg curve peak locations from proton beam therapy experiments using ML approaches. The optimized ML model can provide higher precision in identifying the needed beam dosage for targeted tumor and improving treatment outcomes.

Keywords

• Machine learning
• regression
• proton therapy
• Bragg curve

References

1. Ekinci, F., Bölükdemir, M. H., The effect of the second peak formed in biomaterials used in a slab head phantom on the proton Bragg peak, J. Polytech., 23 (1) 2020, 129-136, http://doi.org/10.2339/politeknik.523001.
2. Ekinci, F., Bostancı, G. E., Dağlı, Ö., Güzel, M. S., Analysis of Bragg curve parameters and lateral straggle for proton and carbon beams, Commun. Fac. Sci.Univ. Ank. Series A2-A3: Phys. Sci. and Eng., 63 (1) (2021), 32-41, https://doi.org/10.33769/aupse.864475.
3. Ekinci, F., Bostanci, E., Güzel, M. S., Dagli, O., Effect of different embolization materials on proton beam stereotactic radiosurgery arteriovenous malformation dose distributions using the Monte Carlo simulation code, J. Radiat. Res. App. Sci., 15 (3) 2022, 191-197, https://doi.org/10.1016/j.jrras.2022.05.011.
4. Gottschalk, B., Proton Therapy Physics, Taylor & Francis Inc., USA, 2012, https://doi.org/10.1201/b22053.
5. Ekinci, F., Bostanci, E., Güzel, M. S., Dağli, Ö., Analysing the effect of a cranium thickness on a Bragg peak range in the proton therapy: a TRIM and GEANT4 based study, St. Petersbg. State Polytech. Univ. J.: Phys. Math., 15 (2) (2022) 64-78, https://doi.org/0.18721/JPM.15207.
6. Carlsson, A. K., Andrea, P. and Brahme, A., Monte Carlo and analytical calculation of computerized treatment plan optimization, Phys. Med. Biol., 42 (1997), 1033-1053, https://doi.org/10.1088/0031-9155/42/6/004.
7. Hall, E. J., Kellerer, A. M., Rossi, H. H., Lam, Y-M.P., The relative biological effectiveness of 160 MeV protons-II, Int. Rad. Onc. Biol. Phys., 4 (1978), 1009-1013, https://doi.org/10.1016/0360-3016(78)90013-5.
8. Lourenço, A., Wellock, N., Thomas, R., Homer, M., Bouchard, H., Kanai, T., MacDougall, N., Royle, G., Palmans, H., Theoretical and experimental characterization of novel water-equivalent plastics in clinical high-energy carbon-ion beams, Physics in Medicine and Biology, 61 (21) (2016), 7623-7638. https://doi.org/10.1088/0031- 9155/61/21/7623.

9. Arib, M., Medjadj, T., Boudouma, Y., Study of the influence of phantom material and size on the calibration of ionization chambers in terms of absorbed dose to water, J. Appl. Clin. Med. Phys., 7 (2006), 55-64, https://doi.org/10.1120/jacmp.v7i3.2264.
10. Samson, D. O., Jafri, M. Z. M., Shukri, A., Hashim, R., Sulaiman, O., Aziz, M. Z. A., Yusof, M. F. M., Measurement of radiation attenuation parameters of modified defatted soy flour-soy protein isolate-based mangrove wood particleboards to be used for CT phantom production, Radiat. Environ. Biophys., 59 (2020), 483-501, https://doi.org/10.1007/s00411-020-00844-z.
11. Kanematsu, N., Koba, Y., Ogata, R., Evaluation of plastic materials for range shifting range compensation and solid phantom dosimetry in carbon-ion radiotherapy, Med. Phys., 40 (2013), 041724, https://doi.org/10.1118/1.4795338.
12. Senirkentli, G. B., Ekinci, F., Bostanci, E., Güzel, M. S., Dağli, Ö., Karim, A. M., Mishra, A., Therapy for mandibula plate phantom, Healthcare, 9 (167) (2021), https://doi.org/10.3390/ healthcare9020167.
13. Ekinci, F., Investigation of tissue equivalence of phantom biomaterials in 4He heavy ion therapy, Radiat. Eff. Defects Solids, 178 (3-4) (2023), 500-509, https://doi.org/10.1080/10420150.2022.2153251.
14. Ekinci, F., Asuroglu, T., Acici, K., Monte Carlo simulation of TRIM algorithm in ceramic biomaterial in proton therapy, Materials, 16 (13) (2023), 4833, https://doi.org/10.3390/ma16134833.
15. Ekinci, F., Bostanci, E., Güzel, M. S., Dagli, Ö., A Monte Carlo study for soft tissue equivalency of potential polymeric biomaterials used in carbon ion radiation therapy, Nucl. Technol., 209 (8) (2023), 1-11, https://doi.org/10.1080/ 00295450.2023.2188144.
16. Borderias-Villarroel, E., et al., Machine learning-based automatic proton therapy planning: Impact of post-processing and dose-mimicking in plan robustness, Med. Phys., 50 (2023), 4480-4490, https://doi.org/10.1002/mp.16408.
17. Lerendegui-Marco, J., et al., Towards machine learning aided real-time range imaging in proton therapy, Sci. Rep., 12 (2022), 2735, https://doi.org/10.1038/s41598-022-06126-6.
18. Chang, C. W., Validation of a deep learning-based material estimation model for Monte Carlo dose calculation in proton therapy, Phys. Med. Biol., 67 (21) (2022), 215004, https://doi.org/10.1088/1361-6560/ac9663.
19. Chen, Y., et al., Understanding machine learning classifier decisions in automated radiotherapy quality assurance, Phys. Med. Biol., 67 (2022), 025001, https://doi.org/10.1088/1361-6560/ac3e0e.
20. Foster, D. G., Artur, E. D., Avarege Neutronic Properties of “Prompt” Fission Products, Los Alamos National Laboraty Report, LA--9168-MS (1982).
21. Ziegler, J. F., SRIM: The stopping and range of ion in matter (2013). Available at: http://www.srim.org/. [Accessed November 2023].
22. Bhat, P., Malaganve, P., Effect of J48 and LMT algorithms to classify movies in the web a comparative approach, Innovations in Computer Science and Engineering. Lecture Notes in Networks and Systems, Springer Singapore, 2021, https://doi.org/10.1007/978-981-33-4543-0_58.
23. Ilyas, H., et al., Chronic kidney disease diagnosis using decision tree algorithms, BMC Nephrol., 22 (1) 2021, 273, https://doi.org/10.1186/s12882-021-02474-z.
24. Rahmayanti, N., Pradani, H., Pahlawan, M., Vinarti, R., Comparison of machine learning algorithms to classify fetal health using cardiotocogram data, Procedia Comp. Sci., 197 (2022), 162-171, https://doi.org/10.1016/j.procs.2021.12.130.
25. Simsekler, M. C. E., Alhashmi, N. H., Azar, E., King, N., Luqman, R., Al Mulla, A., Exploring drivers of patient satisfaction using a random forest algorithm, BMC Med. Inform. Decis. Mak., 21 (1) (2021), 157, https://doi.org/10.1186/s12911-021-01519-5.
26. Açıcı, K., Erdaş, Ç. B., Aşuroğlu, T., Toprak, M. K., Erdem, H., Oğul, H., A random forest method to detect Parkinson’s Disease via gait analysis, Communications in Computer and Information Science, Springer, Switzerland, 2017, https://doi.org/10.1007/978-3-319-65172-9_5.
27. Aşuroğlu, T., Oğul, H., A deep learning approach for sepsis monitoring via severity score estimation, Comput. Methods Programs in Biomed. 198 (2021), 105816, https://doi.org/10.1016/j.cmpb.2020.105816.
28. Oyeleye, M., Chen, T., Titarenko, S., Antoniou, G., A predictive analysis of heart rates using machine learning techniques, Int. J. Environ. Res. Public Health, 19 (2022), 2417, https://doi.org/10.3390/ijerph19042417.
29. Huang, L., Song, T., Jiang, T., Linear regression combined KNN algorithm to identify latent defects for imbalance data of ICs, Microelectron. J., 131 (2023), 105641, https://doi.org/10.1016/j.mejo.2022.105641.
30. Wu, J., et al., Prediction and screening model for products based on fusion regression and XGBoost classification, Comput. Intell. Neurosc., 2022 (2022), https://doi.org/10.1155/2022/4987639.
31. Shin, H., XGBoost regression of the most significant photoplethysmogram features for assessing vascular aging, IEEE J. Biomed. Health Inform., 26 (7) (2022), 3354-3361, https://doi.org/10.1109/JBHI.2022.3151091.
32. Manoharan, A., Begam, K. M., Aparow, V. R., Sooriamoorthy, D., Artificial neural networks, gradient boosting and support vector machines for electric vehicle battery state estimation: A review, J. Energy Storage, 55 (A) (2022), 105384, https://doi.org/10.1016/j.est.2022.105384.
33. Quan, Q., et al., Research on water temperature prediction based on improved support vector regression, Neural Comput. & Applic., 34 (2022), 8501-8510, https://doi.org/10.1007/s00521-020-04836-4.
34. Nilashi, M., Abumalloh, R. A., Minaei-Bidgoli, B., Samad,S., Ismail, M. Y., Alhargan, A., Zogaan, W. A., Predicting Parkinson’s Disease progression: Evaluation of ensemble methods in machine learning, J. Health. Eng., (2022), 2022, https://doi.org/10.1155/2022/2793361.
35. Sharin, S. N., Radzali, M. K., Sani, M. S. A., A network analysis and support vector regression approaches for visualising and predicting the COVID-19 outbreak in Malaysia, Health. Anal., 2 (2022), 100080, https://doi.org/10.1016/j.health.2022.100080.
36. Uddin, S., et al., Comparative performance analysis of K-nearest neighbour (KNN) algorithm and its different variants for disease prediction, Sci. Rep., 12 (2022), 6256, https://doi.org/10.1038/s41598-022-10358x.
37. Lin, G., Lin, A., Gu, D., Using support vector regression and K-nearest neighbors for short-term traffic flow prediction based on maximal information coefficient, Infor. Sci., 608 (2022), 517-531, https://doi.org/10.1016/j.ins.2022.06.090.
38. Fayed, H. A., Atiya A. F., Speed up grid-search for parameter selection of support vector machines, Appl. Soft Comput., 80 (2019), 202-210, https://doi.org/ 10.1016 /j.asoc.2019.03.037.
39. Sun, Y., et al., An improved grid search algorithm to optimize SVR for prediction, Soft Comput., 25 (2021), 5633-5644, https://doi.org/10.1007/s00500-020-05560-w.
40. Hamdia, K. M., Zhuang, X., Rabczuk, T., An efficient optimization approach for designing machine learning models based on genetic algorithm, Neural Comput. & Applic., 33 (2021), 1923-1933, https://doi.org/10.1007/s00521-020-05035-x.
41. Da, L., Sun, K., Random forest solar power forecast based on classification optimization, Energy, 187 (2019), 115940, https://doi.org/10.1016/ j.energy.2019.115940.
42. Chui, K. T., Gupta, B. B., Vasant, P., A genetic algorithm optimized RNN-LSTM model for remaining useful life prediction of turbofan engine, Electronics, 10 (3) (2021), 285, https://doi.org/10.3390/electronics10030285.
43. Beyaz, S., Açıcı, K., Sümer, E., Femoral neck fracture detection in X-ray images using deep learning and genetic algorithm approaches, Jt. Dis. Relat. Surg., 31 (2) (2020), 175-183, https://doi.org/10.5606/ehc.2020.72163.
44. Zhou, J., et al., Employing a genetic algorithm and grey wolf optimizer for optimizing RF models to evaluate soil liquefaction potential, Artif. Intell. Rev., 55 (2022), 5673-5705, https://doi.org/10.1007/s10462-022-10140-5.
45. Lee, Y. G., et al., SHAP value-based feature importance analysis for short-term load forecasting, J. Electr. Eng. Technol., 18 (2023), 579-588, https://doi.org/10.1007/s42835-022-01161-9.
46. Gramegna, A., Giudici, P., Why to buy insurance? An explainable artificial intelligence approach, Risks, 8 (4) (2020), 137, https://doi.org/10.3390/risks8040137.
47. Kim, Y., Kim, Y., Explainable heat-related mortality with random forest and SHapley Additive exPlanations (SHAP) models, Sustain. Cities Soc., 79 (2022), 103677, https://doi.org/10.1016/j.scs.2022.103677.
48. Alenezi, R., Ludwig, S. A., Explainability of cybersecurity threats data using SHAP, 2021 IEEE Symposium Series on Computational Intelligence (SSCI), (2021), 01-10, https://doi.org/10.1109/SSCI50451.2021.9659888.