Evaluation of an automated clinical decision system with deep learning dose prediction and NTCP model for prostate cancer proton therapy.

Objective.To evaluate the feasibility of using a deep learning dose prediction approach to identify patients who could benefit most from proton therapy based on the normal tissue complication probability (NTCP) model.Approach.Two 3D UNets were established to predict photon and proton doses. A dataset of 95 patients with localized prostate cancer was randomly partitioned into 55, 10, and 30 for training, validation, and testing, respectively. We selected NTCP models for late rectum bleeding and acute urinary urgency of grade 2 or higher to quantify the benefit of proton therapy. Propagated uncertainties of predicted ΔNTCPs resulting from the dose prediction errors were calculated. Patient selection accuracies for a single endpoint and a composite evaluation were assessed under different ΔNTCP thresholds.Main results.Our deep learning-based dose prediction technique can reduce the time spent on plan comparison from approximately 2 days to as little as 5 seconds. The expanded uncertainty of predicted ΔNTCPs for rectum and bladder endpoints propagated from the dose prediction error were 0.0042 and 0.0016, respectively, which is less than one-third of the acceptable tolerance. The averaged selection accuracies for rectum bleeding, urinary urgency, and composite evaluation were 90%, 93.5%, and 93.5%, respectively.Significance.Our study demonstrates that deep learning dose prediction and NTCP evaluation scheme could distinguish the NTCP differences between photon and proton treatment modalities. In addition, the dose prediction uncertainty does not significantly influence the decision accuracy of NTCP-based patient selection for proton therapy. Therefore, automated deep learning dose prediction and NTCP evaluation schemes can potentially be used to screen large patient populations and to avoid unnecessary delays in the start of prostate cancer radiotherapy in the future.

Mixed-size spot scanning with a compact large momentum acceptance superconducting (LMA-SC) gantry beamline for proton therapy.

Objective. Lowering treatment costs and improving treatment quality are two primary goals for next-generation proton therapy (PT) facilities. This work will design a compact large momentum acceptance superconducting (LMA-SC) gantry beamline to reduce the footprint and expense of the PT facilities, with a novel mixed-size spot scanning method to improve the sparing of organs at risk (OAR).Approach. For the LMA-SC gantry beamline, the movable energy slit is placed in the middle of the last achromatic bending section, and the beam momentum spread of delivered spots can be easily changed during the treatment. Simultaneously, changing the collimator size can provide spots with various lateral spot sizes. Based on the provided large-size and small-size spot models, the treatment planning with mixed spot scanning is optimized: the interior of the target is irradiated with large-size spots (to cover the uniform-dose interior efficiently), while the peripheral of the target is irradiated with small-size spots (to shape the sharp dose falloff at the peripheral accurately).Main results. The treatment plan with mixed-size spot scanning was evaluated and compared with small and large-size spot scanning for thirteen clinical prostate cases. The mixed-size spot plan had superior target dose homogeneities, better protection of OAR, and better plan robustness than the large-size spot plan. Compared to the small-size spot plan, the mixed-size spot plan had comparable plan quality, better plan robustness, and reduced plan delivery time from 65.9 to 40.0 s.Significance. The compact LMA-SC gantry beamline is proposed with mixed-size spot scanning, with demonstrated footprint reduction and improved plan quality compared to the conventional spot scanning method.

In vivo EPID-based daily treatment error identification for volumetric-modulated arc therapy in head and neck cancers with a hierarchical convolutional neural network: a feasibility study.

We proposed a deep learning approach to classify various error types in daily VMAT treatment of head and neck cancer patients based on EPID dosimetry, which could provide additional information to support clinical decisions for adaptive planning. 146 arcs from 42 head and neck patients were analyzed. Anatomical changes and setup errors were simulated in 17,820 EPID images of 99 arcs obtained from 30 patients using in-house software for model training, validation, and testing. Subsequently, 141 clinical EPID images from 47 arcs belonging to the remaining 12 patients were utilized for clinical testing. The hierarchical convolutional neural network (HCNN) model was trained to classify error types and magnitudes using EPID dose difference maps. Gamma analysis with 3%/2 mm (dose difference/distance to agreement) criteria was also performed. The F1 score, a combination of precision and recall, was utilized to evaluate the performance of the HCNN model and gamma analysis. The adaptive fractioned doses were calculated to verify the HCNN classification results. For error type identification, the overall F1 score of the HCNN model was 0.99 and 0.91 for primary type and subtype identification, respectively. For error magnitude identification, the overall F1 score in the simulation dataset was 0.96 and 0.70 for the HCNN model and gamma analysis, respectively; while the overall F1 score in the clinical dataset was 0.79 and 0.20 for the HCNN model and gamma analysis, respectively. The HCNN model-based EPID dosimetry can identify changes in patient transmission doses and distinguish the treatment error category, which could potentially provide information for head and neck cancer treatment adaption.

Feasibility study of multiple-energy Bragg peak proton FLASH on a superconducting gantry with large momentum acceptance.

BACKGROUND: While the Bragg peak proton beam (BP) is capable of superior target conformity and organs-at-risk sparing than the transmission proton beam (TB), its efficacy in FLASH-RT is hindered by both a slow energy switching process and the beam current. A universal range shifter (URS) can pull back the high-energy proton beam while preserving the beam current. Meanwhile, a superconducting gantry with large momentum acceptance (LMA-SC gantry) enables fast energy switching. PURPOSE: This study explores the feasibility of multiple-energy BP FLASH-RT on the LMA-SC gantry. METHOD AND MATERIALS: A simultaneous dose and spot map optimization algorithm was developed for BP FLASH-RT treatment planning to improve the dose delivery efficiency. The URS was designed to be 0-27 cm thick, with 1 cm per step. BP plans using the URS were optimized using single-field optimization (SFO) and multiple-field optimization (MFO) for ten prostate cancer patients and ten lung cancer patients. The plan delivery parameters, dose, and dose rate metrics of BP plans were compared to those of TB plans using the parameters of the LMA-SC gantry. RESULTS: Compared to TB plans, BP plans significantly reduced MUs by 42.7% (P < 0.001) with SFO and 33.3% (P < 0.001) with MFO for prostate cases. For lung cases, the reduction in MUs was 56.8% (P < 0.001) with SFO and 36.4% (P < 0.001) with MFO. BP plans also outperformed TB plans by reducing mean normal tissue doses. BP-SFO plans achieved a reduction of 56.7% (P < 0.001) for prostate cases and 57.7% (P < 0.001) for lung cases, while BP-MFO plans achieved a reduction of 54.2% (P < 0.001) for the prostate case and 40.0% (P < 0.001) for lung cases. For both TB and BP plans, normal tissues in prostate and lung cases received 100.0% FLASH dose rate coverage (>40 Gy/s). CONCLUSIONS: By utilizing the URS and the LMA-SC gantry, it is possible to perform multiple-energy BP FLASH-RT, resulting in better normal tissue sparing, as compared to TB plans.

Comparison and evaluation of different deep learning models of synthetic CT generation from CBCT for nasopharynx cancer adaptive proton therapy.

BACKGROUND: Cone-beam computed tomography (CBCT) scanning is used for patient setup in image-guided radiotherapy. However, its inaccurate CT numbers limit its applicability in dose calculation and treatment planning. PURPOSE: This study compares four deep learning methods for generating synthetic CT (sCT) to determine which method is more appropriate and offers potential for further clinical exploration in adaptive proton therapy for nasopharynx cancer. METHODS: CBCTs and deformed planning CT (dCT) from 75 patients (60/5/10 for training, validation and testing) were used to compare cycle-consistent Generative Adversarial Network (cycleGAN), Unet, Unet+cycleGAN and conditionalGenerative Adversarial Network (cGAN) for sCT generation. The sCT images generated by each method were evaluated against dCT images using mean absolute error (MAE), structural similarity (SSIM), peak signal-to-noise ratio (PSNR), spatial non-uniformity (SNU) and radial averaging in the frequency domain. In addition, dosimetric accuracy was assessed through gamma analysis, differences in water equivalent thickness (WET), and dose-volume histogram metrics. RESULTS: The cGAN model has demonstrated optimal performance in the four models across various indicators. In terms of image quality under global condition, the average MAE has been reduced to 16.39HU, SSIM has increased to 95.24%, and PSNR has increased to 28.98. Regarding dosimetric accuracy, the gamma passing rate (2%/2 mm) has reached 99.02%, and the WET difference is only 1.28 mm. The D95 value of CTVs coverage and Dmax value of spinal cord, brainstem show no significant differences between dCT and sCT generated by cGAN model. CONCLUSIONS: The cGAN model has been shown to be a more suitable approach for generating sCT using CBCT, considering its characteristics and concepts. The resulting sCT has the potential for application in adaptive proton therapy.

[Epidemic research in the south of Fujian Province in modern times from the viewpoints of overseas communication].

Under the particular geographical environment and social structure, different spatiality of epidemics was observed in the south of Fujian Province. Some important factors cannot be ignored in the study of local epidemics, such as its developed overseas communication, prosperous commercial activities between the East and the West and deep-rooted overseas emigration tradition. In modern times, public health ideas, therapies and prevention measures of west medicine were introduced, taking epidemic disease prevention as a turning point in this area, which promoted medical development of this area objectively, and valuable experience in disease prevention was accumulated.