A machine learning-based model for a dose point kernel calculation

EJNMMI Physics

Original Research

Ignacio Scarinci¹  Pedro Pérez¹  Mauro Valente² 

[1] Instituto de Física Enrique Gaviola (IFEG), CONICET, Av. Medina Allende s/n, 5000, Córdoba, Argentina;Laboratorio de Investigación e Instrumentación en Física Aplicada a la Medicina e Imágenes de Rayos X (LIIFAMIRx), Facultad de Matemática, Astronomía, Física y Computación, Universidad Nacional de Córdoba, Av. Medina Allende s/n, 5000, Córdoba, Argentina;Instituto de Física Enrique Gaviola (IFEG), CONICET, Av. Medina Allende s/n, 5000, Córdoba, Argentina;Laboratorio de Investigación e Instrumentación en Física Aplicada a la Medicina e Imágenes de Rayos X (LIIFAMIRx), Facultad de Matemática, Astronomía, Física y Computación, Universidad Nacional de Córdoba, Av. Medina Allende s/n, 5000, Córdoba, Argentina;Centro de Excelencia en Física e Ingeniería en Salud (CFIS) & Departamento de Ciencias Físicas, Universidad de la Frontera, Avenida Francisco Salazar 01145, 4811230, Temuco, Cautín, Chile;

关键词: Beta emitters;    Dose point kernel;    Internal dosimetry;    Machine learning;   

DOI  :  10.1186/s40658-023-00560-9

 received in 2022-12-28, accepted in 2023-06-13,  发布年份 2023

来源: Springer

【 摘 要 】

PurposeAbsorbed dose calculation by kernel convolution requires the prior determination of dose point kernels (DPK). This study reports on the design, implementation, and test of a multi-target regressor approach to generate the DPKs for monoenergetic sources and a model to obtain DPKs for beta emitters.MethodsDPK for monoenergetic electron sources were calculated using the FLUKA Monte Carlo (MC) code for many materials of clinical interest and initial energies ranging from 10 to 3000 keV. Regressor Chains (RC) with three different coefficients regularization/shrinkage models were used as base regressors. Electron monoenergetic scaled DPKs (sDPKs) were used to assess the corresponding sDPKs for beta emitters typically used in nuclear medicine, which were compared against reference published data. Finally, the beta emitters sDPK were applied to a patient-specific case calculating the Voxel Dose Kernel (VDK) for a hepatic radioembolization treatment with 90\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$^{90}$$\end{document}Y.ResultsThe three trained machine learning models demonstrated a promising capacity to predict the sDPK for both monoenergetic emissions and beta emitters of clinical interest attaining differences lower than 10%\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$10\%$$\end{document} in the mean average percentage error (MAPE) as compared with previous studies. Furthermore, differences lower than 7%\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$7 \%$$\end{document} were obtained for the absorbed dose in patient-specific dosimetry comparing against full stochastic MC calculations.ConclusionAn ML model was developed to assess dosimetry calculations in nuclear medicine. The implemented approach has shown the capacity to accurately predict the sDPK for monoenergetic beta sources in a wide range of energy in different materials. The ML model to calculate the sDPK for beta-emitting radionuclides allowed to obtain VDK useful to achieve reliable patient-specific absorbed dose distributions required short computation times.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

40517_2023_259_Article_IEq62.gif

40517_2023_259_Article_IEq61.gif

Fig. 1

40517_2023_259_Article_IEq58.gif

40517_2023_259_Article_IEq57.gif

40517_2023_259_Article_IEq88.gif

40517_2023_259_Article_IEq87.gif

40517_2023_259_Article_IEq56.gif

40517_2023_259_Article_IEq54.gif

40517_2023_259_Article_IEq53.gif

40517_2023_259_Article_IEq86.gif

40517_2023_259_Article_IEq52.gif

Fig. 4

40517_2023_259_Article_IEq51.gif

Fig. 2

40517_2023_259_Article_IEq90.gif

40517_2023_259_Article_IEq89.gif

40517_2023_259_Article_IEq48.gif

40517_2023_259_Article_IEq85.gif

40517_2023_259_Article_IEq84.gif

40517_2023_259_Article_IEq46.gif

40517_2023_259_Article_IEq45.gif

Fig. 2

40517_2023_259_Article_IEq42.gif

Fig. 1

40517_2023_259_Article_IEq36.gif

40517_2023_259_Article_IEq35.gif

40517_2023_259_Article_IEq32.gif

40517_2023_259_Article_IEq31.gif

40517_2023_259_Article_IEq30.gif

Fig. 1

40517_2023_259_Article_IEq28.gif

40517_2023_259_Article_IEq27.gif

Fig. 2

Fig. 1

42004_2023_919_Article_IEq176.gif

42004_2023_919_Article_IEq175.gif

Fig. 2

Fig. 5

Fig. 8

Fig. 5

42004_2023_919_Article_IEq171.gif

Fig. 4

Fig. 1

42004_2023_919_Article_IEq168.gif

13011_2023_540_Article_IEq40.gif

40517_2023_259_Article_IEq25.gif

Fig. 4

Fig. 1

40517_2023_259_Article_IEq19.gif

40517_2023_259_Article_IEq18.gif

40517_2023_259_Article_IEq17.gif

40517_2023_259_Article_IEq16.gif

40517_2023_259_Article_IEq15.gif

Fig. 3

40517_2023_259_Article_IEq13.gif

Fig. 2

40517_2023_259_Article_IEq11.gif

40517_2023_259_Article_IEq8.gif

40517_2023_259_Article_IEq7.gif

40517_2023_259_Article_IEq6.gif

40517_2023_259_Article_IEq5.gif

13011_2023_540_Article_IEq19.gif

Fig. 2

40517_2023_259_Article_IEq2.gif

40517_2023_259_Article_IEq1.gif

13011_2023_540_Article_IEq14.gif

Fig. 1

13011_2023_540_Article_IEq10.gif

Fig. 7

13011_2023_540_Article_IEq8.gif

13011_2023_540_Article_IEq7.gif

13011_2023_540_Article_IEq6.gif

13011_2023_540_Article_IEq5.gif

Fig. 1

Fig. 1

13011_2023_540_Article_IEq3.gif

Fig. 12

13011_2023_540_Article_IEq1.gif

42004_2023_919_Article_IEq130.gif

Fig. 1

Fig. 11

Fig. 10

Fig. 7

41116_2023_37_Article_IEq177.gif

Fig. 1

41116_2023_37_Article_IEq291.gif

41116_2023_37_Article_IEq290.gif

41116_2023_37_Article_IEq289.gif

41116_2023_37_Article_IEq287.gif

41116_2023_37_Article_IEq286.gif

41116_2023_37_Article_IEq285.gif

41116_2023_37_Article_IEq283.gif

Fig. 1

41116_2023_37_Article_IEq279.gif

41116_2023_37_Article_IEq277.gif

41116_2023_37_Article_IEq275.gif

41116_2023_37_Article_IEq273.gif

41116_2023_37_Article_IEq272.gif

41116_2023_37_Article_IEq271.gif

41116_2023_37_Article_IEq269.gif

41116_2023_37_Article_IEq252.gif

41116_2023_37_Article_IEq251.gif

41116_2023_37_Article_IEq250.gif

41116_2023_37_Article_IEq241.gif

41116_2023_37_Article_IEq228.gif

41116_2023_37_Article_IEq227.gif

41116_2023_37_Article_IEq226.gif

41116_2023_37_Article_IEq225.gif

41116_2023_37_Article_IEq224.gif

41116_2023_37_Article_IEq223.gif

41116_2023_37_Article_IEq222.gif

41116_2023_37_Article_IEq221.gif

41116_2023_37_Article_IEq220.gif

41116_2023_37_Article_IEq219.gif

41116_2023_37_Article_IEq218.gif

41116_2023_37_Article_IEq217.gif

41116_2023_37_Article_IEq216.gif

41116_2023_37_Article_IEq215.gif

41116_2023_37_Article_IEq214.gif

Fig. 3

41116_2023_37_Article_IEq213.gif

41116_2023_37_Article_IEq210.gif

Fig. 4

41116_2023_37_Article_IEq208.gif

Fig. 5

Fig. 3

41116_2023_37_Article_IEq196.gif

41116_2023_37_Article_IEq186.gif

41116_2023_37_Article_IEq184.gif

Fig. 2

41116_2023_37_Article_IEq163.gif

41116_2023_37_Article_IEq160.gif

Fig. 2

41116_2023_37_Article_IEq137.gif

41116_2023_37_Article_IEq132.gif

Fig. 1

41116_2023_37_Article_IEq120.gif