Abstract:574 Objectives: Prompt gamma (PG) imaging technique has been demonstrated to be very promising in real-time proton range monitoring in proton therapy. Despite the high correlation between PG and proton dose (PD) distribution, the detected PG cannot represent the delivered PD and be directly comparable with the planned PD. Here, we propose a deep learning based method to estimate the absolute PD from PG distribution. Methods: Monte Carlo simulation was performed using GATE v8.0 with the physics list QGSP_BERT_HP_EMY. Head and neck CT scans were performed on 14 volunteers, and the data were used to generate the heterogeneous medium distribution in the simulation. For each volunteer, incident monenergistic pencil beams with the energy of 150±0.8 MeV was simulated from two different positions. The spot size of proton beam was 6 mm ×10 mm. The total number of protons for each incidence was 1.2×108. The 3D PD deposited in each volunteer’s brain and the emitted PG with energy ranging from 2 MeV to 8 MeV were stored in a 300×234×300 image with 1 mm ×1 mm ×1 mm voxel size. 31 2D slices were extracted from the each 3D image. A U-net model was implemented to learn the PG-to-PD transform. Integrated CT and simulated PG distribution were used as two input channels for training, while the simulated PD was used as the label. Patches with size of 200×200 were randomly extracted from the images and then flipped both horizontally and vertically to augment the data and avoid over-fitting. L2 was chosen as the loss function to obtain stable solution. 480 epochs with size of 1600 were trained. Each epoch contained 100 batches with mini-batch size of 16. The model was trained with 744 pairs from 12 volunteers and tested with the remaining 2 volunteers. To investigate the performance of this proposed method, the estimated PD was first visually compared with simulated PD. The mean projected range equal the depth of the distal 80% point of the Bragg peak in different slices, D80, was calculated to measure the range accuracy of estimated PD. Also, the relative error of the cumulative dose in 3D ROI was calculated to compare the estimated PD and simulated PD quantitatively by rendering 2D slices together. Results: We observe very similar distribution between estimated PD and simulated PD. The range D80 bias is less than 0.5 mm for volunteer #1, where the Bragg peak locates in homogeneous brain regions. In volunteer #2, when the Bragg peak is near the nasal cavity, the range bias is less than 2.0 mm. The relative error of cumulative dose in a 3D ROI is less than 5% for all cases. Conclusion: The proposed deep learning based method satisfactorily recovers PD distribution from PG and corresponding CT data. This method shows great potential not only for range verification but also for quantitative 3D dose monitoring using PG in proton therapy. Further investigations are under way to implement the polyenergetic pencil beam scanning plans in simulation and estimation. Research Support: The research was supported by the National Natural Science Foundation of China (No. 81727807, No.11575096, No. 11605008) and National Key Research and Development (R&D) Plan of China (Grant ID. 2019YFF0302503 and 2016YFC0105405).

A deconvolution method for estimating proton range from measured intensity distribution of prompt gamma rays

Proton range verification through prompt gamma-ray spectroscopy

3D Prompt Gamma Imaging in a Dual-head Multi-slit System for Proton Beam Range Monitoring

Feasibility Study of Range Verification in Proton Therapy Based on Neutron Information

A full-scale clinical prototype for proton range verification using prompt gamma-ray spectroscopy

Compact Method for Proton Range Verification Based on Coaxial Prompt Gamma-Ray Monitoring: a Theoretical Study

Development of focusing multi-knife-edge camera for proton beam range verification: A Monte Carlo study

Experimental study of range verification of proton radiotherapy using off-line PET/CT

Dose Reconstruction with Compton Camera During Proton Therapy Via Subset-Driven Origin Ensemble and Double Evolutionary Algorithm

Feasibility Study of an Integrated Prompt Gamma Imaging and PET System for Range Verification in Proton Therapy

Deep Learning Based 3D Dose Estimation from Prompt Gamma Distribution for Proton Therapy Monitoring

3D prompt gamma imaging for proton beam range verification

Processing of prompt gamma-ray timing data for proton range measurements at a clinical beam delivery

Value and pitfalls of the lateral lung scan.

Fast range measurement of spot scanning proton beams using a volumetric liquid scintillator detector

Imaging of prompt gamma rays emitted during delivery of clinical proton beams with a Compton camera: feasibility studies for range verification

Machine learning approach for proton range verification using real-time prompt gamma imaging with Compton cameras: addressing the total deposited energy information gap

The NOVO project neutron detection for real-time range verification in proton therapy, A Monte Carlo feasibility study

Investigation of intra-fractionated range guided adaptive proton therapy: (Part I) on-line PET imaging and range measurement

Investigation of time-resolved proton radiography using x-ray flat-panel imaging system

Hybrid in-beam PET- and Compton prompt-gamma imaging aimed at enhanced proton-range verification