A comparative study of machine-learning approaches in proton radiography using energy-resolved dose function. (February 2023)

• Record Type: Journal Article
• Authors: B.G., Alaka; Bentefour, El H.; Teo, Boon-Keng Kevin; Samuel, Deepak
• Is Part Of: Physica medica. Volume 106(2023)
• Publication Date: 2023-02
• Subjects: Proton radiography -- Machine learning -- Energy-resolved dose function; Medical physics -- Periodicals; Biophysics -- Periodicals; Biophysics -- Periodicals; Imagerie médicale -- Périodiques; Radiothérapie -- Périodiques; Rayons X -- Sécurité -- Mesures -- Périodiques; Physique -- Périodiques; Médecine -- Périodiques; 610.153
• DOI: 10.1016/j.ejmp.2023.102525
• Languages: English
• ISSNs: 1120-1797

Abstract

Abstract: Purpose: The feasibility of machine learning (ML) techniques and their performance compared to the conventional χ 2 -minimization technique in the context of the proton energy-resolved dose imaging method are presented. Materials and method: Various geometries resembling a wedge and varying gradients are simulated in GATE to obtain energy-resolved dose functions (ERDF) from proton beams of different energies. These ERDFs are used to predict the WEPL using a conventional technique and other ML-based methods. The results are compared to gain an understanding of the performance of ML models in proton radiography. Results: The results obtained from the χ 2 -minimization technique indicate that it is robust and more reliable compared to the ML-based techniques. It is also observed that the ML-based techniques did not mitigate the effect of range-mixing but seem to be more affected by it compared to the χ 2 -minimization technique. Substantial data reduction was required in order to make the results of ML-based methods comparable to that of χ 2 -minimization. We also note that such data reduction might not be possible in a clinical setting. The only advantage in using the ML-based technique is the computational time required to generate a WEPL map which, in our case study, is 10-30 times shorter than the time required for the conventional χ 2 -minimization technique. Conclusions: The first results from this preliminary study indicate that the ML techniques failed to be on par with the conventional χ 2 -minimization technique in terms of the achievable accuracy in the predictions of WEPL and in the mitigation of range-mixing effects in the WEPL image. Modern strategies like the GAN-based models may be suitable for such applications. Highlights: Implementation of three machine-learning models to generate proton radiographs Implementation of three machine-learning models to generate proton radiographs Features and labels are created using proton energy-resolved dose imaging technique The first results show that ML models may not reduce range-mixing effects significantly The computational time required to generate radiographs is within two mins This is faster by a factor of 10 compared to the conventional technique