Three‐dimensional dose and LETD prediction in proton therapy using artificial neural networks

Abstract Purpose Challenges in proton therapy include identifying patients most likely to benefit; ensuring consistent, high‐quality plans as its adoption becomes more widespread; and recognizing biological uncertainties that may be related to increased relative biologic effectiveness driven by linear energy transfer (LET). Knowledge‐based planning (KBP) is a domain that may help to address all three. Methods Artificial neural networks were trained using 117 unique treatment plans and associated dose and dose‐weighted LET (LET D ) distributions. The data set was split into training ( n = 82), validation ( n = 17), and test ( n = 18) sets. Model performance was evaluated on the test set using dose‐ and LET D ‐volume metrics in the clinical target volume (CTV) and nearby organs at risk and Dice similarity coefficients (DSC) comparing predicted and planned isodose lines at 50%, 75%, and 95% of the prescription dose. Results Dose‐volume metrics significantly differed ( α = 0.05) between predicted and planned dose distributions in only one dose‐volume metric, D 2% to the CTV. The maximum observed root mean square (RMS) difference between corresponding metrics was 4.3 Gy RBE (8% of prescription) for D 1cc to optic chiasm. DSC were 0.90, 0.93, and 0.88 for the 50%, 75%, and 95% isodose lines, respectively. LET D ‐volume metrics significantly differed in all but one metric, L 0.1cc of the brainstem. The maximum observed difference in RMS differences for LET D metrics was 1.0 keV/μm for L 0.1cc to brainstem. Conclusions We have devised the first three‐dimensional dose and LET D ‐prediction model for cranial proton radiation therapy has been developed. Dose accuracy compared favorably with that of previously published models in other treatment sites. The agreement in LET D supports future investigations with biological doses in mind to enable the full potential of KBP in proton therapy.