Highlights
- •XGBoost was trained to predict the gamma passing rate for patient-specific QA.
- •The machine learning model was evaluated in a multicentric scenario.
- •The model provided accurate and conservative estimations of the gamma passing rate.
- •Machine learning can streamline the radiotherapy workflow.
- •Machine learning models should be verified within centers before clinical use.
Abstract
Purpose
Patient-specific quality assurance (PSQA) is performed to ensure that modulated treatment
plans can be delivered as intended, but constitutes a substantial workload that could
slow down the radiotherapy process and delay the start of clinical treatments. In
this study, we investigated a machine learning (ML) tree-based ensemble model to predict
the gamma passing rate (GPR) for volumetric modulated arc therapy (VMAT) plans.
Materials and methods
5622 VMAT plans from multiple treatment sites were selected from a database of Institution
1 and the ML model trained using 19 metrics. PSQA analyses were performed automatically
using criteria 3%/1 mm (global normalization, absolute dose, 10% threshold) and 95%
action limit. Model’s performance was evaluated on an out-of-sample test set of Institution
1 and on two independent sets of measurements collected at Institution 2 and Institution
3. Mean absolute error (MAE), as well as the model’s sensitivity and specificity,
were computed.
Results
The model obtained a MAE of 2.33%, 2.54% and 3.91% for the three Institutions, with
a specificity of 0.90, 0.90 and 0.68, and a sensitivity of 0.61, 0.25, and 0.55, respectively.
Small positive median values of the residuals (i.e., the difference between measurements
and predictions) were observed for each Institution (0.95%, 1.66%, and 3.42%). Thus,
the model’s predictions were, on average, close to the real values and provided a
conservative estimation of the GPR.
Conclusions
ML models can be integrated into clinical practice to streamline the radiotherapy
workflow, but they should be center-specific or thoroughly verified within centers
before clinical use.
Keywords
To read this article in full you will need to make a payment
Purchase one-time access:
Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online accessOne-time access price info
- For academic or personal research use, select 'Academic and Personal'
- For corporate R&D use, select 'Corporate R&D Professionals'
Subscribe:
Subscribe to Physica Medica: European Journal of Medical PhysicsAlready a print subscriber? Claim online access
Already an online subscriber? Sign in
Register: Create an account
Institutional Access: Sign in to ScienceDirect
References
- Tolerance limits and methodologies for IMRT measurement-based verification QA: Recommendations of AAPM Task Group No. 218.Med Phys. 2018; 45: e53-e83https://doi.org/10.1002/mp.12810
- A technique for the quantitative evaluation of dose distributions.Med Phys. 1998; 25: 656-661https://doi.org/10.1118/1.598248
- A new metric for assessing IMRT modulation complexity and plan deliverability.Med Phys. 2010; 37: 505-515https://doi.org/10.1118/1.3276775
- Impact of plan parameters on the dosimetric accuracy of volumetric modulated arc therapy: plan parameters and VMAT dosimetric accuracy.Med Phys. 2013; 40: 071718
- Modulation indices for volumetric modulated arc therapy.Phys Med Biol. 2014; 59: 7315-7340https://doi.org/10.1088/0031-9155/59/23/7315
- The effect of MLC speed and acceleration on the plan delivery accuracy of VMAT.BJR. 2015; 88: 20140698https://doi.org/10.1259/bjr.20140698
- On the need for tuning the dosimetric leaf gap for stereotactic treatment plans in the Eclipse treatment planning system.J Appl Clin Med Phys. 2019; 20: 68-77https://doi.org/10.1002/acm2.12656
- Use of metrics to quantify IMRT and VMAT treatment plan complexity: a systematic review and perspectives.Phys Med. 2019; 64: 98-108
- Complexity metrics for IMRT and VMAT plans: a review of current literature and applications.BJR. 2019; 92: 20190270https://doi.org/10.1259/bjr.20190270
- Improvement of prediction and classification performance for gamma passing rate by using plan complexity and dosiomics features.Radiother Oncol. 2020; 153: 250-257
- Machine learning for patient-specific quality assurance of VMAT: prediction and classification accuracy.Int J Radiat Oncol Biol Phys. 2019; 105: 893-902
- A mathematical framework for virtual IMRT QA using machine learning: Virtual IMRT QA.Med Phys. 2016; 43: 4323-4334https://doi.org/10.1118/1.4953835
- IMRT QA using machine learning: A multi-institutional validation.J Appl Clin Med Phys. 2017; 18: 279-284https://doi.org/10.1002/acm2.12161
- Predicting gamma passing rates for portal dosimetry-based IMRT QA using machine learning.Med Phys. 2019; 46: 4666-4675
- Application and comparison of machine learning models for predicting quality assurance outcomes in radiation therapy treatment planning.Inf Med Unlocked. 2020; 18100292https://doi.org/10.1016/j.imu.2020.100292
- Systematic method for a deep learning-based prediction model for gamma evaluation in patient-specific quality assurance of volumetric modulated arc therapy.Med Phys. 2021; 48: 1003-1018
- Deep nets vs expert designed features in medical physics: an IMRT QA case study.Med Phys. 2018; 45: 2672-2680
- Multi-task autoencoder based classification-regression model for patient-specific VMAT QA.Phys Med Biol. 2020; 65: 235023
- A deep learning-based prediction model for gamma evaluation in patient-specific quality assurance.Med Phys. 2018; 45: 4055-4065
- Quality assurance of RapidArc in clinical practice using portal dosimetry.BJR. 2011; 84: 534-545
- Survey of patient-specific quality assurance practice for IMRT and VMAT.J Appl Clin Med Phys. 2021; 22: 155-164
- National survey of patient specific IMRT quality assurance in China.Radiat Oncol. 2019; 14: 69https://doi.org/10.1186/s13014-019-1273-5
- Multi-centre audit of VMAT planning and pre-treatment verification.Radiother Oncol. 2017; 124: 302-310
- Treatment plan complexity metrics for predicting IMRT pre-treatment quality assurance results.Australas Phys Eng Sci Med. 2014; 37: 475-482
- Quantification of beam complexity in intensity-modulated radiation therapy treatment plans: IMRT beam complexity.Med Phys. 2014; 41: 021716
- Penalization of aperture complexity in inversely planned volumetric modulated arc therapy: penalization of aperture complexity in inversely planned VMAT.Med Phys. 2012; 39: 7160-7170https://doi.org/10.1118/1.4762566
Chen T, Guestrin C. XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco California USA: ACM; 2016, p. 785–94. https://doi.org/10.1145/2939672.2939785.
- Tabular data: Deep learning is not all you need.Inform Fusion. 2022; 81: 84-90https://doi.org/10.1016/j.inffus.2021.11.011
- From local explanations to global understanding with explainable AI for trees.Nat Mach Intell. 2020; 2: 56-67
- Report of AAPM Task Group 219 on independent calculation-based dose/MU verification for IMRT.Med Phys. 2021; 48https://doi.org/10.1002/mp.15069
- Detailed evaluation of Mobius3D dose calculation accuracy for volumetric-modulated arc therapy.Phys Med. 2020; 74: 125-132
- Commissioning and clinical implementation of Mobius3D and MobiusFX: experience on multiple linear accelerators.Phys Med. 2020; 80: 1-9
- Commissioning and performance evaluation of RadCalc for the Elekta unity MRI-linac.J Appl Clin Med Phys. 2019; 20: 54-62
- Evaluation and comparison of second-check monitor unit calculation software with Pinnacle 3 treatment planning system.Phys Med. 2018; 45: 186-191https://doi.org/10.1016/j.ejmp.2017.12.004
- Parallel/Opposed: IMRT QA using treatment log files is superior to conventional measurement-based method.J Appl Clin Med Phys. 2015; 16: 4-7https://doi.org/10.1120/jacmp.v16i1.5385
- Challenges in translational machine learning.Hum Genet. 2022; 141: 1451-1466
- Commissioning of the varian TrueBeam linear accelerator: a multi-institutional study: Multi-institutional commissioning of five TrueBeam linear accelerators.Med Phys. 2013; 40: 031719
- Technical Note: Multicenter study of TrueBeam FFF beams with a new stereotactic diode: can a common small field signal ratio curve be defined?: Multicenter TrueBeam FFF study with a new diode.Med Phys. 2016; 43: 5570-5576
- Golden beam data provided by linear accelerator manufacturers should be used in the commissioning of treatment planning systems.Phys Eng Sci Med. 2022; 45: 407-411https://doi.org/10.1007/s13246-022-01134-2
- Reference dataset of users’ photon beam modeling parameters for the eclipse, pinnacle, and raystation treatment planning systems.Med Phys. 2020; 47: 282-288https://doi.org/10.1002/mp.13892
Article info
Publication history
Published online: April 25, 2023
Accepted:
April 14,
2023
Received in revised form:
March 2,
2023
Received:
November 24,
2022
Identification
Copyright
© 2023 Associazione Italiana di Fisica Medica e Sanitaria. Published by Elsevier Ltd. All rights reserved.
